AHEAD OF PRINT

The survival soil on which the traditional martial arts culture depends is becoming thinner and thinner. A large number of superb martial arts skills and martial arts experience disappear with the passing of the older generation of martial artists. In order to realize the information preservation of traditional Wushu, this paper puts forward the establishment and application of traditional Wushu intelligent learning resource database under the background of big data. In the context of big data, data mining technology is used to realize personalized recommended learning services, obtain the feature structure of intelligent learning system through operation, extract the average dynamic features of specific data in the resource database, obtain fuzzy constraints according to the value, determine the membership function, and adjust the membership function parameters through training fuzzy neural network, so as to gradually improve the reasoning accuracy. Finally, the matching rule set is used to filter the resource data packets to achieve the purpose of communication. In order to verify the effect of the model, compared with the traditional model, the results show that when the number of nodes is 500, the average transmission rate is as high as 90%, the average delay is 12 seconds, and the throughput performance is 91%. It can be verified that the model designed in this paper can effectively improve the propagation rate, reduce the average delay and strengthen the throughput performance.

A physical model of a tall space livestock house was established using Prosim software and the relevant boundary conditions were set under the replacement of ventilation conditions. The exhaust air volume calculated by different ventilation times is simulated to the effort of the indoor pollutant concentration distribution and the measuring points are evenly arranged in the space. The study found that the replacement ventilation effect is optimal and the indoor pollutant concentration is the lowest when the indoor air supply and exhaust air volume are increased to 45 times/h.

Chebyshev collocation scheme and Finite difference method plays central roles for solving fractional differential equations (FDE). Therefore purpose of this paper is to solve fractional mathematical problem of diffusion by Chebyshev collocation method which turns the original problems into the system of fractional ordinary differential and algebraic equations by imposing orthogonality property. This system is solved by implementing Finite difference method. The numerical illustrations confirm that the combination of these two methods allow us to establish one of the best truncated solution in the series form.

As the latest product of the innovative integration of information technology, big data has had a huge impact on many fields. Big data technology has changed the traditional audit method, expanded the scope and field of auditing, and injected vitality into the national big data audit work. Full audit coverage has given social auditing a new mission and posed new challenges to creating mega data audit platforms. Driven by the overall goal of full audit coverage, the construction and innovation of a big data audit platform are particularly important. This subject combines data-related technologies and uses more mature data mining algorithms to try to establish an audit platform, and analyzes the three aspects of requirements, concepts and strategies for building an audit platform, dividing it into five modules: data center, acquisition, pre-processing, analysis and data visualization. The research results show that by predicting the revenue indicators of Company A, the overall error of the prediction results of the big data audit platform is controlled at 0.79%, which is better than 8.42% of the traditional audit model; by scoring the two audit models, the score of the big data audit method is 0.7024 higher than that of the traditional model, which is 0.3933, indicating that its accuracy, efficiency comprehensiveness, sustainability, and low risk are In this way, this paper explores a new effective audit path and method, which provides a good basis for the enterprise’s “industry price, quality and trust payment” and guides the way forward for future mega Data audits.

With the progress of computer technology, discipline construction, new management ideas, management modes and management technologies gradually penetrate into all aspects of information construction, which leads to the further improvement of the level of informationisation. To better realise the dual functions of discipline construction and comprehensive management of graduate education, the advantages of the Hopfield Network (HN) algorithm, which is easy to realise and improve information processing efficiency and programme maintenance, are brought into play. This paper conducts an in-depth study on the comprehensive management system of graduate education based on the HN algorithm. The technical feasibility of the integrated postgraduate education management system is 73%. The economic feasibility of the integrated graduate education management system is 68%, with other feasibility variables remaining unchanged. Moreover, the design of the integrated graduate education management system is more reasonable at 71%, which is much higher than the 29% for the traditional system.

To address the problem of inaccurate positioning of new energy vehicles during cooperative detection, this paper investigates the positioning accuracy of different combinations of radar and infrared sensors. In order to meet the control requirements of the vehicle co-detection formation, a distributed control method considering the vehicle dynamics characteristics is adopted for this purpose. And considering the cost of sensors, the optimal parameters are calculated, and an optimal aircraft co-detection configuration design scheme based on the SCP model (structure-behavior-performance model) is proposed. At the working level of multi-craft cooperative detection guidance, a two-layer cooperative guidance structure is designed based on the basic framework. Research on vehicle configuration design, vehicle data transmission, processing technology, and cooperative detection and guidance is carried out. The basic principles of the particle swarm optimization algorithm and the algorithmic flow of the discrete particle swarm algorithm are applied to calculate the assignment of targets by the particle swarm optimization algorithm. The GDOP values are analyzed according to the simulation results, and it can be concluded that: the change rate of 2~5 sensors accuracy can reach 71.1%, 71.6%, and 71.5%, respectively, when the infrared sensor spacing is 1km~3km. 5~10 sensors accuracy change rate is only 53.3%, 52.2%, and 49.7%, respectively, which shows that the value is improved by 20%, indicating that the collaborative sensing accuracy of the aircraft has been effectively improved, which is of great help to the collaborative detection of new energy vehicles based on SCP model.

Based on the accurate application of precision teaching can enable students to conduct efficient learning and achieve the educational goal of all-round development. In this paper, a big data precision teaching model is constructed to stage the teaching process, set the data corresponding to each stage as random variables, and apply the AIC criterion to reflect the complexity of the model. Then the evaluation model of the precision teaching algorithm is constructed by calculating the test set error based on cross-validation to find the precision teaching solution under random variables. In the last stage, it is known from the observation data of the students that the optimal decision function is derived from the variable selection, according to which the targeted development ideas of precision teaching are obtained. The experimental results showed that the practice subjects were divided into control and experimental groups, follow-up tests were conducted, and the percentage of failures in the control group was 4%, while the percentage of failures in the experimental group was 9%. The scores of the experimental class were higher when analyzed by the comprehensive attainment degree. It shows that the development and application of the platform under precision teaching can not only make up for some problems and shortcomings of the traditional classroom but also enable students to get better learning results in this precision teaching process.

Party building and ideological and political education have received much attention in recent years, and the construction of its work practice platform is particularly important. In order to improve the level of party building and ideological and political education in colleges and universities, In this paper is based on the information fusion technique to fuse relevant factors after data, feature, and decision levels. Using sampling calculation, the percentage of relevant factors was calculated as 37%, 38%, and 25%, and the Kalman filter algorithm was used to test the information data. The two factors that accounted for more were selected as the platform construction indexes, and the information was fused with the maximum likelihood estimation and D-S algorithm to build the party construction and ideological and political education work practice platform. It is evident that the practice platform based on information fusion technology can promote the development of ideological and political education.

The purpose of this paper is to provide a fast, efficient, and low-cost digital design method for the design of suction mufflers in refrigerator compressor plants. The main content of this paper is to establish a three-dimensional acoustic model of the compressor suction muffler by using Solidedge software, combine the upper suction port of the muffler with the valve group, then use Hypermesh to mesh the acoustic model and carry out the theoretical calculation, transform the control differential equation in the solution domain into the integral equation on the boundary by Green's function, discretize the boundary and carry out the numerical solution. The boundary is discretized and solved numerically. The transfer matrix quadrupole parameters are used to describe the acoustic components, and the transfer matrix of the discontinuous cross section is obtained to transform the complex system into a chain superposition of simple systems. The dependent variable in the partial differential equation to be solved is changed to a linear system of equations, and the transmission loss is calculated from the incident and transmitted sound pressure. The theoretical calculation of the transmission loss of the muffler is performed using LMS. Virtual. Lab software and the design solution is continuously optimized using this software. The experimental analysis shows that the designed muffler is printed and trimmed by 3D printing equipment and loaded onto the transmission loss test bench for experimental testing, and the highest theoretical value differs from the test result by 10dB.

Air data computer simulation system is a platform for studying aircraft air data system. It can also replace real air data computer and can be used in the experiment of ground integrated avionics system, which greatly reduces the cost and risk of the experiment. However, when the simulation system replaces the real system, whether its reliability, integrity and other indicators can meet the requirements becomes the key to the problem. Based on computer simulation theory and simulation credibility evaluation theory, this paper designs and implements two types of atmospheric data computer simulation systems and evaluates the credibility of the simulation systems. The basic theory of computer simulation and the general process of simulation are expounded, the evaluation theory of simulation credibility is introduced, and the evaluation calculation method of simulation credibility is given. The results show that the highest reliability of the reliability evaluation using the method in this paper reaches 90%, the highest error rate is 0.06%, and the average accuracy, efficiency and complexity of the hundred experiments are 97.31% and 97.00% and 0.154%. The highest reliability of the subjective comprehensive evaluation algorithm for reliability evaluation is 65%, the highest error rate is 0.27%, and the average accuracy, efficiency and complexity of the hundred experiments are 70.59% and 69.74%, and 0.502%. Therefore, the method in this paper not only has high reliability, low error rate, but also has strong comprehensive effectiveness.

In order to solve the problem that a large number of low-value and easy-to-consume graphic designs consume the human and financial resources of the society and also waste the inherent value of design resources, by dividing the layers of pixels, elements, relations, planes and applications, this article defines the corresponding feature representations and studies related methods for quantifying geometric features, perceptual features and style features. The content includes the extraction method of element colour, the calculation method of layout-aware feature and colour-match-aware feature, and the pairwise comparison method of style feature. The experimental results show that the method proposed in this article has lower complexity and is significantly better than other algorithms. Comparing the method proposed in this article with other methods, the average time of this algorithm is 2.6 s under the condition of guaranteed images. The study of the design feature-driven graphic image style calculation method is not only of great significance for solving repeated design problems but also provides a basis and reference for the future development of intelligent design.

Teachers are the leaders of teaching activities, and the quality of teachers is the core of their ability to effectively carry out teaching practice. The quality of teachers’ team is closely related to the educational environment. This paper first aims at the origin and development of the concept of literacy construction and believes that the literacy of teachers’ team in the educational environment is an extension of literacy, and it is also a necessary literacy for teaching. First, it is essential to carry out the calculation of the literacy matching weight. Second, the construction of teacher team literacy mainly includes teachers’ professional literacy, teacher teaching literacy, teacher information literacy, teacher learning literacy and teacher personal characteristics. The construction of team literacy is inseparable from the influence of environmental factors, mainly including the needs of school students, school innovation atmosphere, school education environment and other educational environmental factors. Combined with the quality of teachers’ team, a set of teacher team quality construction framework based on the educational environment is designed. At the same time, the use of information technology innovation and breakthroughs, combined with school teacher team literacy construction requires certain educational environment support, and it is necessary to provide targeted training for literacy construction, create a suitable teaching atmosphere, get rid of traditional teaching presuppositions, and start from the educational environment. The problems that appear in the process are used as a basis to improve the construction of literacy according to the problems. Finally, using deep learning to predict and analyse the data, the experimental results show that the accurate prediction can reach 96.89%, indicating that various factors in the educational environment will affect the construction of teacher team literacy.

With the rapid development of China’s electric power industry, the cost management of electric power projects, which is closely related to the power grid transmission and substation projects, has become more and more important. Do a good investment feasibility analysis of power projects. Optimize power engineering design. Reasonable determination of power project cost and effective control of power project cost are the main problems to be solved by the project cost management of power grid transmission and transformation projects. In this paper, DCN-BIM model is established based on DCN algorithm as the theory. A detailed calculation and analysis of the influencing factors and stage share of the cost management of power grid transmission and substation projects is carried out. Through calculation, it can be seen that the share of power products in the project cost is about 15.43% due to their special characteristics. The lowest cost share in the project cost is the project tender. Its share is only 10.03%. The highest cost share is for equipment materials. Its cost share is as high as 51.34%. The design phase accounted for 23% of the cost. The lowest cost share in the bidding phase was only 21%.

In recent years, digital technology has rapidly penetrated various industries and brought various opportunities for developing these industries. Based on two digital technologies, BP neural network and the Gray Wolf algorithm, this article analyzes the application of digital technology in environmental art design at Xi’an Eurasian National University, taking environmental art design as an example. The article uses these two algorithms to analyze the site and design principles of the campus, followed by detailed design and consideration of the campus environment design, mainly from greenery design, groundscape and road pavement design, and campus landscape lighting design. The analysis of the collected data in this paper shows that students’ satisfaction with the central teaching area is generally low, at around 20%, and the satisfaction with the northwest teaching area is the highest, all above 40%, up to 61%. The students’ satisfaction with the plant landscape on the campus was generally at a medium level. However, the satisfaction level for loblolly pine is not very high, and overall is around 30%, while the satisfaction level for the cedar is high, and everyone’s satisfaction is higher than 60%. Therefore, this paper takes the students’ satisfaction with the campus landscape environment into consideration in the landscape design, which is conducive to achieving the perfect integration of landscape art design and ecological environment and achieving the realistic goal of both aesthetic and environmental protection.

In order to explore the stability of resource and environmental economics model, the author proposes a research on the stability of resource and environmental economics model based on hyperbolic sine function. The method recommends key technical problems and solutions based on information represented by hyperbolic sine functions, and explores the research on the stability of resource and environmental economics models. Research has shown that, the research on the stability of the resource and environmental economics model based on the hyperbolic sine function is about 21% more efficient than the traditional method. By ensuring the smooth acceptance of spillover technologies in traditional industries, it will stimulate a wave of innovation in the whole society. This can not only promote economic development, but also have important significance for human survival and protection of natural resources.

With the rapid development of the times, my country has made good innovation and progress in the Internet of Things technology, which has brought a lot of technical guarantees for computer network security. In the process of computer network security analysis and research, it is necessary to introduce the Internet of Things technology. In this paper, the CNN-GRU-PSO network is used to optimize the information classification control. Compared with the traditional model method, the CNN-GRU-PSO model method improves the accuracy rate from the original 86.4% to the original 95.5%. Nearly 10%; the precision rate increased from 84.3% to the original 91.2%, and the accuracy rate increased to nearly 7%; the recall rate was increased from 86.4% to the original 93.5%, and the accuracy rate increased to nearly 9%. The CNNGRU-PSO model optimizes network security management and formulates strict prevention mechanisms to ensure that computer networks can operate efficiently and safely.

With the rapid upgrade of AI in today’s era, the technology of designing garments by merely taking measurements of women’s clothing through traditional pattern makers followed by hand-drawn clothing style patterns by clothing designers has become increasingly inadequate to meet people’s needs. In this paper, the study of super-resolution reconstruction technology according to deep learning improves the resolution of women’s clothing style images and helps to save production costs. The overall clothing style migration method, according to the optimization DCGAN algorithm, improves the efficiency of intelligent design of women’s clothing. It is shown that the reconstruction time of a clothing style image can be completed in only 0.26s, which is much smaller than the reconstruction time of the SRCNN convolutional neural network algorithm, which is 4.30s. The loss value of the live network is 0.79, the loss value of the classical network is 1.23, and the loss value of the optimization algorithm is 0.48. The algorithm in this paper has the smallest training loss value. Therefore, the use of deep learning technology can solve the problem of traditional women’s apparel design, relying on the designer’s experience and inspiration and generating design solutions intelligently.

For the current artwork, 3D printing design simulation technology has the disadvantage of low clarity and poor effect, which cannot effectively describe the designer's work. To this end, this paper investigates the artistic aesthetic and cultural value orientation of 3D printing. First, the 3D digitization work of the oil painting was carried out by using the multi-baseline photogrammetry method in photogrammetry technology. Then the non-measurement digital camera was checked for internal orientation elements, the control panel was photographed, and the internal orientation elements of the camera were measured using the spatial backward rendezvous solution method. Next, the digital image acquisition of the oil painting is carried out, and finally, the photos are sequentially introduced into the Lens photo multi-baseline photogrammetry system to generate the color point cloud model of the oil painting after performing the steps of homonymous point selection null-three matching and beam leveling. In order to enhance the visualization effect, the color point cloud model was used to generate a digital model of the TIN irregular triangular network to realize the 3D digitization of oil painting art. The results of the study show that the thickness of each layer obtained by UV inkjet printing of five types of inks: yellow, green, blue, black, and white is about 11.19 μm, 12.71 μm, 7.91 μm, 9.47 μm, and 13.12 μm, respectively. 3D printing technology for positive impact on art design needs to be recognized and respected, and the better development of related technologies can make the advantages and advancement of the technology itself more fully embodied.

Facial expressions can reflect people’s inner emotions to a certain extent, and studying facial expressions can help psychologists capture expression information in time and understand patients’ psychological changes quickly. In this paper, we establish a multi-channel convolutional neural network face expression recognition model based on the fusion of the attention mechanism. With the help of the attention mechanism and multi-channel convolutional neural network, we input expression images and perform average pooling and maximum pooling, output the features with high recognition after pooling, and identify the features with high recognition in expression images throughout the process. And with the help of multi-scale feature fusion, we improve the detection of subtle changes, such as the corners of the mouth and the eyes of the expression image target. The loss function is used to calculate the loss rate of facial expression images, which leads to the correct rate of facial expression recognition by a multi-channel convolutional neural network based on the fusion of attention mechanisms. It is demonstrated that the highest recognition correct rate of the multi-channel convolutional neural network faces expression recognition model with attention mechanism fusion is 93.56% on the FER2013 dataset, which is higher than that of the MHBP model by 23.2%. The highest correct recognition rate on the RAF-DB dataset is 91.34%, which is higher than the SR-VGG19 model by 19.39%. This shows that the multi-channel convolutional neural network face expression recognition based on the fusion of attention mechanisms improves the correct rate of facial expression recognition, which is beneficial to the research and development of psychology.

Public health events are sudden, public in nature and have serious social hazards. The COVID-19 outbreak coincided with the Lunar New Year, which had a direct or indirect impact on all areas of society. Previous studies related to emergencies have found that a considerable number of college students lacked experience in dealing with emergencies, were not emotionally stable enough, lacked analysis and decision-making ability, were easily suggestible and acted more impulsively. Therefore, in this paper, based on the existing actual information, combined with the awareness and understanding of college students’ mental health, and based on the existing research results, the Hopfield-mental health model is used as a theoretical basis to study the trend of changes in college students’ mental health. The results of the study show that 83.21% of the people are more concerned about the situation of this new crown pneumonia epidemic and they think that the new crown epidemic has seriously affected their living habits; 65.45% thought that this new crown pneumonia epidemic did not have any major impact on their school life. The five sources of psychological stress, including academic, employment, economic, interpersonal relationship and love, were calculated and analysed in the model, which showed that employment stress, academic stress and economic stress were the largest sources of psychological stress among college students in this new pneumonia epidemic, accounting for 89%, 81% and 93%, respectively. They were followed by interpersonal and romantic stress, with 31% and 52%, respectively.

With the development of smart grids, power grids have accumulated massive amounts of data in various links such as power generation, transmission, substation, distribution, power consumption, and dispatch. More and more big data applications are beginning to be applied in various professional fields of the power grid. Promote the application and value discovery of smart grid big data through data fusion inside and outside the grid. Grid data has become an important asset for enterprise development, but power grid enterprises lack effective technical means to solve the whole life cycle monitoring and relationship of power grid data assets. Aiming at the relationship between power grid data assets, this paper proposes a set of grid data asset relationship and intelligent classification framework that integrates knowledge graph and Internet of Things. First, the grid knowledge graph extraction relationship is carried out by ProjE algorithm. Then, the relationship between power grid data assets and intelligent classification framework that integrates knowledge graph and Internet is proposed. Finally, the corresponding classification application is proposed by using intelligent classification algorithm. Experimental results show that the intelligent classification accuracy rate can reach 93.12% under the relationship between the knowledge graph and the Internet data assets, which has a new idea for the future development of the relationship between power grid data assets.

With the increasing importance of music information retrieval, the construction of effective music recognition methods has gradually become one of the research focuses. The vocal map features of the collected music pronunciation signals are extracted. The research in this paper is primarily based on the basic physical characteristics of music notes; once these characteristics are identified, then they are mathematically extracted and analysed (voiceprint feature method), and a recognition model is established. On this basis, the process of adaptive separation and recognition of music signals is completed. Finally, the performance of different recognition types is verified and evaluated through experiments. The results show that the average accuracy rate of the modified algorithm within a certain range reaches 73.6%; additionally, the average accuracy rate is increased by 10.95% compared with the audio recognition based on Internet of Things (IoT) data, and is more accurate than the audio recognition method based on data collection, showing an improvement of 20.75%. This shows that the modified recognition algorithm adopted in the present research has stable and high accuracy for comprehensive music types with different characteristics. Finally, the identification method proposed in this paper shortens the time by 85.71% compared with the identification method of data collection, and shortens the identification time by 83.33% compared with the IoT identification method. This greatly improves the recognition of different musical feature types.

With the rapid entry into deep mining of coal mines in China, the impact composite dynamic disaster of thick and hard layer mines and mine earthquakes has increasingly become a major disaster that threatens the safe and efficient mining of deep coal. Studying its occurrence mechanism, disaster prevention and control system has become a new major scientific issue in the field of coal mine safety. This paper proposes a framework for a composite dynamic disaster prevention and control system framework for thick and hard rock mines. First, the gravity forms, extents and deformation characteristics of different rock layers of the structural model are analysed, and the expressions of concentrated force and periodic breaking step distance of rock beams in thick and hard rock layers at the fixed support end are deduced. Then, according to the cause of the shock composite dynamic disaster, finally, the specific testing and calculation methods of mine earthquakes in thick hard rock mines are designed. The regional and local measures to manage compound dynamic disasters are put forward. Experiments show that the system is successfully applied to the mining practice of working face, and the results of water conservancy and stress monitoring support the rationality of the system. And through the implementation of impact prevention and control measures, the safety and disaster prevention and control of the working face was finally realised.

The stability and high quality of electricity are the basic factors which ensure that the residents and enterprises lead a happy and productive life. Therefore, in order to meet the requirements of residents’ life and enterprise production, it is necessary to improve the efficiency and accuracy of power grid fault diagnosis. In this paper, the knowledge graph is integrated into the power grid fault diagnosis, and the fault diagnosis system of the knowledge graph is constructed to realise the fault diagnosis of the power grid. The article first completes the knowledge graph construction through knowledge extraction, knowledge fusion and knowledge processing; then, it completes the construction of the fault scheduling knowledge graph through power grid equipment fault records, entity attribute extraction, coreference resolution, relation extraction, relation screening and data integration; finally, combined with the fault information knowledge analysis technology, it builds a power grid fault diagnosis system using a knowledge graph. Experiments show that using the system to diagnose the fault quantity, fault location and fault analysis information of the pilot power grid not only has ideal efficiency but also has high accuracy.

With the development of socialist market economy, enterprises are faced with an increasingly complex development environment. In the context of socialist market economy, enterprises can only obtain better development if they continuously promote the improvement of internal control and management system, and effectively carry out financial risk control and management. This paper discusses the procedures and construction measures of the internal control and management system of the enterprise from the requirements and contents of the enterprise financial risk management, using the F-score model. We hope to strengthen the exploration of enterprise financial risk management measures by further improving the enterprise’s internal control system, so as to further enhance the quality of enterprise financial risk management, effectively predict, prevent, control and respond to the relevant financial risks of the enterprise, and build a stable development path for the enterprise.

In the past two years, with the development of big data technology and the revolutionary innovations it has caused within various industries, there has been a boom in the use of big data to transform and optimize government workflow within the field of public administration. In this paper, we build a theoretical framework through government statistics, focusing on the four elements of government statistics. The decision tree algorithm is used to analyze the indicators of the government statistical work problem, mainly from three aspects: grassroots statistical institutions, government revenue, and government forestry construction statistics. The accuracy of the decision tree algorithm is controlled between 83% and 91%, with a variation range value of 8%, which is more accurate than the other two algorithms. The decision tree algorithm helps improve the efficiency and quality of government statistics, thus contributing to the processing of statistical work problems and reform strategies. This study is important for deepening the reform of the Chinese government administration system, improving the construction of a service-oriented government, ensuring the quality of statistical information, and guiding the benign development of statistical work.

In order to construct the morphological model of Chinese excellent calisthenics athletes, overcome the shortage of evaluation of single and multiple indexes in the study of the morphology of calisthenics athletes in the past, the author proposes a sports science model research based on fractional differential equation. Sports biomechanics, as an independent discipline within sports science, the general task of the study of motion biomechanics is to evaluate the effect of force on perfectly achieving a given goal in the process of interaction between biological system and external environment, the author takes the outstanding male calisthenics athletes of Chinese college students as the research object, and adopts the method of literature and mathematical statistics, the morphological indexes were analyzed and studied, and the morphological model was established, through factor analysis, the morphology of Chinese outstanding male college student aerobics athletes is divided into four factors: Body fullness factor, limb scale factor, body width factor, body circumference factor, the weights of the four factors are 0.36, 0.31, 0.17 and 0.15, respectively. Chinese outstanding male college student aerobics athletes have the morphological characteristics of medium height and well-developed upper arm and lower limb muscles.

Among all kinds of coal production disasters, the consequences of gas disaster are the most serious. As the existing coal mine gas explosion disaster pre-control management theory and method system is not satisfactory, the neural Turing machine (NTM) deep learning network algorithm is used to calculate and analyse the risk source early warning identification of coal mine gas explosion accidents. Institute with data sets of gas gas accident knowledge base matter each event to cause an (basic or intermediate events) as an example, through the study of the depth of NTM network algorithm calculation analysis shows that self-rescuer failure, personnel peccancy operation, such as downhole safety management does not reach the designated position is easy to cause important hazard of gas explosion accident, the probability to cause an 0.567. Based on the constructed NTM deep learning network algorithm, the risk factors and their weights in the early warning identification of gas explosion accidents are calculated and analysed. Through calculation analysis, it can be seen that the highest weight of risk factors is gas concentration, with a weight of 96. In the early warning identification of hazard sources, the hazard factor next to gas concentration is mine combustibles, with a weight of 75.

In the multimedia context, it is important to enrich the teaching forms, challenge the traditional teaching concepts and realize the innovation of education mode. In this paper, a detailed review of translation strategies for college students in the multimedia context is presented, and the traditional GLR translation teaching analysis algorithm is analyzed. To compensate for the shortcomings of low translation teaching efficiency caused by over-fitting in the traditional GLR translation teaching analysis algorithm, a Bayesian model is constructed, and an adversarial neural network is built on its basis. Generate a translation teaching innovation model applicable to the translation teaching of university students. The translation teaching method is evaluated using the BLEU evaluation method. Experimental results: Both the correct translation rate of utterances based on the statistical computing method and dynamic memory algorithm reached 90%-95%. The traditional GLR translation teaching analysis algorithm achieved 95% correctness in recognizing declarative sentences, while the correctness rate for question and exclamation sentences was less than 95%. The correct translation rate of all the statements of the innovative model of translation teaching reached more than 97%. It can be seen that: The innovative model of translation teaching for college students with multimedia backgrounds is simpler and faster in calculation and more practical than other translation teaching algorithms, which is suitable for English translation work of college students and meets the proofreading needs of college students for translation teaching.

In the context of the prosperity of the Internet economy and the progress of the cultural industry, the Internet cultural industry is expanding in a big way based on the economic development of the cultural industry, supported by the Internet technology platform and serving the cultural and spiritual experience of consumers. The Internet cultural industry is gradually becoming the backbone of the country’s comprehensive strength for the amount. The Internet to boost the development of China’s cultural industry and cultural export to the outside has become a general trend. In 2018 China’s Internet cultural industry added value accounted for the first time exceeded 3 trillion of GDP, and the added value of the Internet cultural industry in 2021 further increased to 6.63% of GDP. Thus, it can be seen that the share of Wen Internet culture industry in China’s GDP is a steady upward trend and the future development is clear. This paper uses the vector autoregressive model to model and analyze the cultural consumption data of Internet users, and the analysis results show that in 2021, China’s Internet cultural consumption mainly tends to music and video, games, literature, animation, and online cultural content information services in five areas, among which games account for 36.52% of the total Internet cultural consumption. Due to the improvement of China’s comprehensive strength, the difference in CPI of Internet culture consumption between urban and rural groups in China is 11, and the gap is gradually narrowing, and the countryside will be a big new market in the future. The Internet group culture user group as a whole tends to be younger, with the student group accounting for 26% of the user group. Based on the analysis of this study, it is concluded that Internet cultural consumption tends to be young, cultural innovation is the main driving force to promote consumption, and future development is unstoppable.

The development of computer maintenance has now become the focus of attention in the education sector, standing in the overall situation of economic, social and educational development, to comprehensively improve the level of computer maintenance education, and teaching quality has become an important measure to meet the new round of technological revolution and industrial change. The traditional teaching model has not been able to adapt to the industry and enterprise demand for job competence, with the development of VR technology computer maintenance training platform. This study uses the NTM-VR model as the theoretical basis and incorporates the data of traditional practical training platform and VR practical training platform into the model for analysis and calculation. The calculation results show that the acceptance of the VR practical training platform by college students is quite high at 67.4%, much higher than 32.6% of the traditional platform. And under the uniform learning intensity, the learning outcome based on the VR computer maintenance practical training platform is 76%, much higher than that based on the traditional computer maintenance practical training platform of 24%.

Multi-exposure image fusion as a technical means to bridge the dynamic range gap between real scenes and image acquisition devices, which makes the fused images better quality and more realistic and vivid simulation of real scenes, has been widely concerned by scholars from various countries. In order to improve the adaptive fusion effect of multi-exposure images, this paper proposes a fusion algorithm based on multilayer perceptron (MLP) based on the perceptron model and verifies the feasibility of the algorithm by the peak signal-to-noise ratio (PSNR), correlation coefficient (PCC), structural similarity (SSMI) and HDR-VDR-2, an evaluation index of HDR image quality. Comparison with other algorithms revealed that the average PSNR of the MLP algorithm improved by 4.43% over the Ma algorithm, 7.88% over the Vanmail algorithm, 10.30% over the FMMR algorithm, 11.19% over the PMF algorithm, and 11.19% over the PMF algorithm. For PCC, the MLP algorithm improves by 20.14%, 17.46%, 2.31%, 11.24%, and 15.36% over the other algorithms in that order. For SSMI, the MLP algorithm improved by 16.99%, 8.96%, 17.17%, 14.41%, and 4.85% over the other algorithms, in that order. For HDR-VDR-2, the MLP algorithm improved by 3.02%, 2.79%, 6.84%, 4.90%, and 6.55% over the other algorithms, in that order. The results show that the MLP algorithm can avoid image artifacts while retaining more details. The MLP-based adaptive fusion method is a step further in the theoretical study of multi-exposure image fusion, which is of great significance for subsequent research and practical application by related technology vendors.

With the improvement of information technology, all industries have developed rapidly. However, the education industry is lagging behind in the process of informationization. Most still follow the educational paradigm of the past, where teachers rely on personal experience to make judgments and teaching decisions for students, just like a blind man feeling like an elephant. In this paper, we design a teaching quality monitoring and evaluation strategy according to the K-modes algorithm in big data technology, which is a dynamic data collection and intelligent analysis based on the learning process and learning effectiveness, ensuring data authenticity and also realizing multi-dimensional data collection and multi-angle evaluation and analysis, and finally comparing each index of three different algorithms. It is verified that the minimum error sum of squares of the optimized K-modes algorithm is 711, and the correct rate is 0.97, while the values of this metric for the other two algorithms are 1587 and 986, and the correct rates are 0.91 and 0.92, respectively. Therefore, the designed mechanism is a system with a superior evaluation effect and a feasible prediction model for faculty development in local vocational colleges.

English education for college students is a major instructional programme with significant developmental importance; yet, its conversations are also crucial. Reflexivity in conversational repair refers to the tendency of speakers to go back to the beginning of the repairing component rather than to implement the repair directly after the wrong word. In order to help listeners solve the connection problem, speakers use different back-referencing strategies. In this study, we used a quantitative and qualitative approach to explore the patterns and strategies of finger-back repair in Chinese college students’ English conversations and to compare their similarities and differences with native speakers in the use of finger-back onset words. The study found that the proportion of self-repetition was the highest among college students, reaching about 75%, and the proportion of reorganisation and insertion strategies was very small, about 23%, which reflected the characteristics of college students’ ability in linguistic information processing and online processing; moreover, there were significant differences between them and native speakers in the use of reflexive priming words. This paper borrows from the attention theory of cognitive psychology to explain the results and to point out the implications of this study for the teaching of spoken English in college.

To meet the needs of small and medium-sized processing plants by establishing a software and hardware platform for a robotic arm control system with a hierarchical structure and designing a robot robotic arm motion control system that is adaptable to multiple environments and cost-effective. In this paper, the industrial robot robotic arm motion control system is designed based on the consistent fusion tracking algorithm under the decentralized Internet. While using the VisualC++ development environment for writing human-machine interface programs using the MFC framework, the device controls the robotic arm body to complete predetermined movements or operational tasks based on command information, sensing information, and the robot controller. The AMS1086CD-3.3 low-voltage differential linear regulator chip used in this design effectively implements the circuit design from 4.8V to 3.4V and successfully meets the 3.4V voltage supply required by the controller. The industrial robot robotic arm control system adopts an adaptive backstepping algorithm controller, and the error of the tracking system at 0 seconds in the simulation results is always kept within the range of [-1~1] that meets the requirements, which effectively reduces the control error of the robotic arm position. Therefore, the robotic arm control studied in this paper is composed of a complete control system in hardware and software, and the designed adaptive backstepping algorithm controller achieves the demand for good control performance of the robotic arm proper.

The construction industry is booming, but construction projects will consume much non-renewable energy and cause environmental pollution. In response to this problem, this article establishes a low-carbon architectural design model and discusses the application of renewable energy decorative materials in modern architectural design in detail. Previous studies have shown that renewable energy decorative materials are widely used in ventilation design, thermal insulation design, interior design and lighting design, and the weights of carbon emissions in the use stage are 20%, 46%, 29% and 33%, respectively. In 2021, the usage weights of renewable energy decorative materials in ventilation design, thermal insulation design, interior design and lighting design will be 70%, 72%, 80% and 70%, respectively, far exceeding the usage of traditional materials.

Based on the lack of management level and other problems in enterprise management innovation, this paper analyzes the innovation pathway of enterprise management by logistic regression algorithm. Firstly, the management pathway is set as a data set, and the feature vector of the model is defined, the probability model of logistic regression is derived from the input data vector, and the model defines the likelihood probability of the training set. Then the log-likelihood function of the logistic regression model with beta distribution is calculated, the feature weights of the data set are obtained using this function, and finally, the effective countermeasures of enterprise management innovation pathways under logistic regression are obtained according to these weights. The results show that by examining the variance comparison of different enterprise properties, private and sole proprietorship enterprises have the highest scores in the external organizational management factors, 3.86 and 3.72, respectively. The factors in the strategic technological capability have the lowest scores compared with other factors. It can be seen that the construction of comprehensive innovation management capability of enterprises is a long-term accumulation process, and the logistic regression algorithm helps enterprises to develop feasible innovation management pathway strategies according to their capabilities.

The background of big data has developed deeply, the application of the field has been broadened, and the value of data has been vigorously manifested. In order to study the interaction between physical culture education and exercise in universities in this era, this paper uses Clementine 12.0 data mining software to build a data mining model of association rules of university physical culture education courses and mine the course feature vectors. Based on the mining results, we designed the second classroom physical culture education courses with different physical culture characteristics. Constructing a scoring method and rating scale for the effectiveness of physical exercise among college students, and the physical exercise index scores are obtained through fuzzy operations. Finally, the interaction between physical culture education and exercise in universities in this context is analyzed according to the relationship between physical culture education courses and physical exercise performance. After the physical culture education course began, the physical exercise intensity score of the experimental group of first-year college girls increased by 8%, the physical exercise time score increased by 10%, the physical exercise frequency score increased by 15.2%, and the total physical exercise score increased by 7% after the physical culture education course. This shows that university physical culture education is positively correlated with college students’ physical activity, and campus physical culture has a significant predictive effect on students’ subjective performance of physical activity behavior. Optimizing university physical culture education not only improves students’ physical quality and promotes the development of their physical and mental health but also provides a reference for strengthening students’ physical education.

To verify the feasibility of robust speech recognition based on deep learning in sports game review. In this paper, a robust speech recognition model is built based on the generative adversarial network GAN algorithm according to the deep learning model. And the loss function, optimization function and noise reduction front-end are introduced in the model to achieve the optimization of speech extraction features through denoising process to ensure that accurate speech review data can be derived even in the game scene under noisy environment. Finally, the experiments are conducted to verify the four directions of the model algorithm by comparing the speech features MFCC, FBANK and WAVE. The experimental results show that the speech recognition model trained by the GSDNet model algorithm can reach 89% accuracy, 56.24% reduction of auxiliary speech recognition word error rate, 92.61% accuracy of speech feature extraction, about 62.19% reduction of training sample data volume, and 94.75% improvement of speech recognition performance in the speech recognition task under noisy environment. It shows that the robust speech recognition based on deep learning can be applied to sports game reviews, and also can provide accurate voice review information from the noisy sports game scene, and also broaden the application area for deep learning models.

Big data has become an important reference and helper tool for enterprise efficiency improvement. Effective data mining and analysis can be utilized by enterprises to enhance the user experience of products or to develop products and services based on user needs analysis. In order to explore the application of data analytics, this study focuses on big data analytics from the cloud computing perspective by reviewing massive references and online data, as well as fieldwork visits to specific situations. The results reflect that data analytics is used in various industries to enhance the benefits of enterprises. The hotel management, Internet fields, new media industry and education industry examined in this paper are all helpful in improving efficiency. After the utilization of the data analytics technology, the market size of China’s cross-border direct big data analytics technology broadcast e-commerce is on the rise from 2020 to 2021, and the annual market size of big data analytics technology is expected to go beyond 100 billion yuan in 2022, a growth rate of 210% compared to the same period last year. Thus, data analytics technology will bring endless social and economic benefits. How to reasonably use data is a problem that needs to be seriously considered.

With the rapid development of modern information technology, especially the continuous improvement of computer network technology, the application of education management system in teaching is becoming more and more extensive. Therefore, education management system and machine learning will become an important combination direction of education. First, design and implement a complete network education management system based on B/S architecture, and design from the overall system design, detailed design and database design. Among them, the computer language combined with the SQL Server database realizes the network teaching function and the education system management function. Then, PSO-SVM machine learning is adopted to make personalized learning course recommendation for students. Multi-dimensional data analysis and feature extraction. Finally, the PSO-SVM proposed in this paper is applied to the education management system for modeling training, and compared with other traditional machine learning personalized recommendation accuracy and likeness of learning course recommendation. The experimental results show that the PSO-SVM proposed in this paper is superior to other traditional machine learning models in terms of personalized learning course recommendation and favorability, with an accuracy rate of 94.7%.

Modern 21st-century society is the cultivation of practical ability, innovation ability, and comprehensive ability of talents, and the cultivation of innovative talents has put forward higher requirements for experimental teaching. With the development of artificial intelligence, “Internet+” and other technologies, the legal education model is facing significant challenges and opportunities in experimental teaching, and the traditional teaching model cannot meet its needs in the experimental teaching of law using “Internet+,” artificial intelligence, and other technologies. The design and implementation of the virtual experimental teaching platform of law can hit the pain points of legal documents, case seminars, mock courts, and legal clinics, etc. The experimental platform driven by information technology and the traditional law experimental courses is mutually supportive. The teaching of law experiment center makes up for the shortage of traditional experimental teaching mode, and the system integrity, functional automation, teacher-student interactivity, and ease of use of the LETS experiment platform, all of which can greatly promote the effectiveness of teaching law courses.

In order to improve the ideological and political education of college students, this paper constructs a cloud technology education platform based on big data analysis and collection technology. Firstly, the collaborative filtering algorithm is used to filter and collect student information; secondly, the Pearson correlation formula is used for pre-processing, MAE (mean absolute error value) is used as an evaluation index; and finally, the data is combined with FCM (fuzzy C-mean) algorithm to filter and analyze as needed. Its application performance is examined to verify the rationality and practicality of the construction of the cloud technology education platform. The analysis results show that, compared with the three-tier neural network education platform and MOOC (Massive Open Online Course) education platform, the educational resource transmission time of the cloud technology education platform constructed in this paper is reduced by 1/3, and the load balancing deviation is reduced by 1/4. Among four different servers, the average error value of the platform constructed in this paper is as low as 0.35% and can reach as high as 0.65%. The detection rate reaches 97.64%, the false alarm rate is only 2.93%, and the leakage rate is only 1.13%. It can be seen that the cloud technology education platform constructed in this paper can improve the utilization rate of educational resources and realize the sharing of educational resources.

English is the most widely used language in the world. For Chinese students, fluent English will have a positive impact on their future study and work. However, with the progress and development of science and technology, the traditional college English education model can no longer meet the needs of the current social development, so it is necessary to use advanced teaching methods such as information technology, intelligence and big data to improve the teaching quality. The inclusion of blockchain technology in the category of local innovation is the official start of the central government’s promotion for the development and application of blockchain technology in China. Under this background, big data, as a new technology, can effectively improve the teaching efficiency of college English and promote the comprehensive development of college students. This paper analyses the problems existing in college English teaching and discusses the innovative strategies of college English teaching combined with the wide application of big data.

The trajectory tracking of badminton players’ arm shots can be used effectively to enhance the player’s shot quality. To track the image trajectory of the batting arm, it is necessary to calculate the body posture ratio and tightness of the target area of the batting arm, to filter the background interference of the image segment of the batting and to complete the efficient tracking of the trajectory of the arm batting image. The traditional method combines the adaptive threshold segmentation method to extract the hitting arm target from the background, but ignores filtering out the background interference of the hitting image fragment. This paper proposes a trajectory tracking method based on the morphological operator of the batter image of the arm stroke. The method consists of (1) differentially calculating the image sequence of the hitting arm during two consecutive shots, (2) estimating the Gaussian model parameters of the differential image of the hitting arm during the hitting process, (3) extracting the outline of the moving target of the hitting arm during the hitting process and then calculating the body posture ratio of the hitting target area and compactness, (4) filtering the background interference of the shot image fragment, (6) constructing a global matching approximation function of the moving target and (7) finally determining the motion trajectory of the badminton arm of the batter. Simulation results show that the proposed method can effectively track the target of the hitting arm during the hitting process and generate a continuous trajectory of the hitting arm.

The extended Kadomtsev-Petviashvili (eKP) equation in fluids is investigated by using Lie symmetry analysis. The symmetry reductions, together with the simplest equation method, are used to obtain exact solutions of the eKP equation. Finally, we derive the conservation laws of eKP by using multiplier approach.

Chinese rural ecotourism is of great significance to promote rural revitalization. As Chinese consumers’ requirements for rural ecotourism gradually increase, seeking the integration of regional characteristics of pastoral culture elements is the top priorities of Chinese rural ecotourism attractions. The creative design of pastoral culture is a breakthrough point in creating characteristic rural ecotourism attractions and truly developing China’s rural economy. Based on this, this paper sorts out the current situation of China’s rural ecotourism attractions, and puts forward feasible suggestions for the design and improvement of future pastoral culture creativity in China’s rural ecotourism.

Character styling design can clearly show the background of story characters and the characteristics of the times in the performance of stage plays. Integrating traditional culture with the art of stage plays is important for developing theatrical communication. In this paper, we analyze the factors that impact theatrical communication in the context of big data. Based on the original innovation diffusion model, it analyzes the limitations of its application, analyzes the innovation characteristics of theatrical stage makeup modeling from a qualitative perspective, finds that its diffusion characteristics do not conform to the prerequisite assumptions of the original innovation diffusion model, and confirms the improvement direction of the innovation diffusion model. Based on the analysis of audience data by the full data analysis method, the main influencing factors affecting the diffusion of opera heritage are identified, and their practical significance in the improved model is analyzed. The original innovation diffusion model is improved quantitatively, and an iterative diffusion model is established. Empirical analysis of the iterative diffusion model was conducted using the actual diffusion data of opera stage makeup styling. The research results show that the initial diffusion rates of the products are, in descending order, Cheese Superman, TikTok, Watermelon Video, and Punchbowl. Among them, the cumulative diffusion of TikTok is the highest at 14, and the diffusion rate of Watermelon Video is 0.68. It indicates that the above products effectively spread opera culture and highlight the charm of opera stage makeup styling.

In order to study the waterproof performance of elastic rubber gasket in shield tunnel lining joints, an innovative sensitivity analysis method is proposed by combining the Monte Carlo method with the stochastic finite element method (FEM) in this paper. The sensitivity values of the waterproof performance respecting to elastic rubber gaskets are obtained via the ANSYS Probabilistic Design System (PDS) module, in which the parameters of material hardness, coordinates of the hole center, apertures are selected as random input variables. Meantime, the extent of the tolerance effect of the random parameters on the waterproof performance is explored.

For the third-party cross-border e-commerce sharing platform, due to the lack of management and operation mechanism, the current sharing platform has relatively high fees, insufficient publicity, and consumption methods are not conducive to Chinese consumers. There are obvious problems such as insufficient personnel training. In this article, we use the CGA-LSO-BP network to solve the problems of various systems disjoint, complicated department settings, confusing distribution of powers and responsibilities in the training base, as well as unclear division of team functions, technology mismatch, and the training base itself. The scale, social reputation and other factors lead to the difficulty of reducing financing channels and other related operation and management issues to study and analyze. The results show that the minimum error can reach 0.5% for CGA-LSO-BP, which is much smaller than the traditional algorithm. It is proved that the algorithm can help the BP neural network to jump out of the local optimal value to a certain extent, and play an active role in the regression task. In addition, the improved CGA-LSO-BP neural network based on this can provide a good reference for various problems in cross-border e-commerce, such as disjointed systems, complicated department settings, and chaotic distribution of rights and responsibilities, and propose optimal solutions.

The paper establishes a related differential equation model about changes in financial interest rates. It uses information related to liquidity to feedback the law and stability of differential equations in interest rate changes. The article applies stochastic processes and partial differential equations to complex financial networks to confirm node yields in financial market networks. It confirms the existence of interest rate stickiness in Chinese financial markets. The advantage of this interest rate model is that when the external economic environment changes, the state of interest rates will also change accordingly.

In order to enhance national competitiveness while ensuring the rational operation of world trade activities, the highly internationalized development of cobalt trade is promoted. This paper analyzes the current situation of world cobalt trade through the collection and collation of trade network data. That is, each node on the network represents an activity subject, and the relationship between nodes and nodes exists in terms of commodity demand and supply, reflecting the network Spatio-temporal divergence characteristics of global commodity trade. By measuring the topological network indicators such as density, point degree, intermediary centrality, and proximity to the center, the influence of a country in the trade network and whether it occupies the central position can be more clearly defined, effectively capturing the strength of the relationship between the nodes in the world trade network and facilitating the understanding of the evolution mechanism of the world trade network, and using the QAP regression model to calculate the relationship and trade volume between two countries that affect the trade connection. The results of the study show that China’s cobalt production in 2021 is 2,105 tons, down 8.5% year-on-year. With the passage of time, the overall scale of the international trade network of bauxite is growing, the number of countries involved in the trade is increasing, the trade volume is growing, and the trade relationship between countries is gradually increasing, which reflects the good development trend of global bauxite trade.

Diamond, a wide bandgap semiconductor material, has excellent physicochemical properties. It has great potential for application in high temperature, high frequency, high power electronic devices and other high technology fields. In order to study the electronic properties of diamond more precisely, an AM-response surface model is developed in this paper to investigate the electronic structures of diamond, P-doped diamond and N-doped diamond surfaces in depth. It is shown that there are three forms of charge states in the single vacancy on the diamond surface. When E=0 V, the negative charge energy level is -0.5 mV, the positive charge is 1 mV, and the zero level remains 0. And its energy level is unstable. In contrast, the double vacancy charge on the diamond surface varies depending on the valence band taken by E, and only one charge state exists. When E<0, the diamond surface vacant electron nature is negative charge state. When E>0, it is positive charge energy level. The electronic properties of the P-doped diamond semiconductor material are calculated to have a constant positive charge (1 mV). The electronic property of N-doped diamond semiconductor material is constant negative charge (-1mV).

Because of the current problems of cultural and creative tourism in tourism destination management under the cultural theory perspective, this paper aims to explore the new development direction of cultural and creative tourism. This paper uses the gray-scale correlation method model to construct a correlation degree model, analyze the factors tourists consider in choosing tourist destinations, determine the weights of each influencing factor, and extract the main influencing factors. On this basis, combined with a questionnaire survey method, we analyze the market demand for cultural and creative tourism, tourists’ satisfaction, and the existing problems of the cultural and creative tourism market and put forward suggestions for the problems. In terms of market demand, the highest percentage of tourists’ motivation for tourism is to experience different cultures 60.55%, followed by relaxation 57.35% and appreciation of scenery 55.4%, which shows that the cultural tourism market has great potential; however, there are still many problems in the current cultural tourism, the most prominent being excessive commercialization and loss of cultural authenticity, which accounts for 75%, followed by high ticket prices, lack of creativity in products, however, the service level needs to be improved to 40%, indicating that all these problems need to be paid attention to. Cultural and creative tourism is a form of tourism preferred by tourists, and many problems still need to be further explored.

As one of the important feature categories in urban geographic data, buildings are the key thematic elements to be represented in large-scale urban mapping with the high speed of urban digital construction. The identification and extraction of buildings are of great significance for feature extraction, feature matching, image interpretation and mapping. However, the great variability of building size, shape, color, orientation, etc., in remote sensing images poses a great challenge to building detection. To this end, this paper proposes an algorithm based on multi-feature multi-scale fusion for the automatic extraction of buildings in remote sensing images are represented in the form of roofs. It is difficult to represent all buildings with a single feature because of the different colors, textures and shapes of building roofs. Effective features to describe buildings are proposed, including edge density and edge distribution, brightness contrast, color contrast and other features to describe building edge brightness. We propose effective features to describe buildings, including edge density and edge distribution, luminance contrast, color contrast and other underlying features to describe the edges, luminance and color of buildings, and adding special structural features such as main direction orthogonality and target integrity and symmetry to describe buildings by multiple features together.

Moreover, the K-value nearest neighbor classification algorithm is used to train a series of samples, and the weights of each feature in the multi-feature model are obtained through iterative learning to obtain the multi-feature linear model and calculate the visual saliency of buildings in the sliding window; finally, the proposed algorithm has experimented with several groups of high-resolution remote sensing images respectively, and the multi-scale multi-feature fusion model algorithm is used as the Erkoff random field model to compare the algorithm. The results of this paper show that the proposed multiscale multi-feature fusion model algorithm improves by 10.82% for building classification accuracy extraction and 13.96% for feature selection extraction accuracy, and finally, the comparison from the shape optimization effect figure concludes that the multiscale multi-feature fusion model can achieve better extraction accuracy and practical effect for buildings in remote sensing images, which has certain practicality and It has certain practicality and superiority. It promotes the in-depth application of multi-feature multi-scale combined high-resolution remote sensing image-building extraction in geographic states, road traffic and other industries.t

With the rapid development of the Internet, security issues are becoming more and more prominent, and since most information is transmitted through the Internet today, Internet security is particularly important. When the Internet was designed, only mutual compatibility and interoperability between networks were considered, and security issues were not fully considered. As a result, as the Internet continues to grow, security issues are becoming more and more serious. One of the more difficult attacks is the Distributed Denial of Service (DDoS) attack, which has many forms of attacks, is harmful, and is difficult to identify and defend. Therefore, building a global Internet security protection system to achieve effective protection against DDoS attacks is the main work of this research paper. In this paper, we propose an artificial intelligence DDoS attack protection system, which implements a controller and switch auto-detection model by extending the protocol and establishing an optimization model to realize a low-load and low-latency traffic monitoring scheme; for DDoS attacks. We propose the attack inspection algorithm SCVAE based on Variational Encoder (VAE) and Spectral Clustering. in order to mitigate DDoS attack traffic, the protection system uses the QoS traffic control method, builds the application flow hierarchy model, and filters the attack traffic endured by the system by setting the application flow bandwidth limit as well as the traffic priority dual policy. Finally, a Mininet-based simulation test environment is built to evaluate the model, and different test indexes are set for different system modules to evaluate their actual performance. The results of this paper show that in the network traffic monitoring test, the artificial intelligence DDoS attack protection algorithm can respond to the attack more quickly by reducing the average 73ms per sampling compared with other algorithms; in the attack traffic identification test, the comparison accuracy (P) is improved by 15.14%, the accuracy (AC) is improved by 13.26%, the recall (R) is reduced by 9.23%, and the F1 measurement criteria improved by 23%. The test verifies that the artificial intelligence DDoS attack protection system can achieve real-time monitoring of each performance parameter and also illustrates the feasibility and practicality of the research content of this paper, which strengthens the construction of the technical means of Internet security protection and further enhances the Internet security defense capability.

With the development of information technology, the news media industry has entered the all-media era. Today, everyone is a communicator, and the audience has more diversified channels to obtain news information, but it also leads to the truly useful news being easily overwhelmed. In order to analyze the strategy of news dissemination in the all-media era, this paper analyzes different types of news media platforms and audiences based on big data mining technology. The mining results show that cultural news has the largest share on radio, TV, and video sites, with 27.5% and 57.3%, respectively. Social news had the largest share on news sites, at 29.8%. Political news has the largest share of mobile clients at 27.9%. Economic news has the largest share on social media platforms, with 23.5%. In addition, the news is viewed much more on new media, such as video websites and social platforms, than on traditional media, such as newspapers, magazines, TV and radio. In terms of the exposure rates of different types of news audiences to media platforms, the average exposure rates of newspapers and magazines, TV and radio, news sites, mobile clients, social platforms and video sites are 8.83%, 28.30%, 44.49%, 59.57%, 71.03% and 90.46%, respectively. In the era of full media, news dissemination should focus on applying new communication technologies, enriching the presentation form of news, and selecting reasonable topics for the characteristics of audiences on different platforms. The analysis of news communication strategy based on big data mining can grasp the pain points in current news communication, which is of great guidance for transforming news communication in the era of full media.

As an emerging business model and economic form, the sharing economy is a new trend to promote global economic growth in the future. Public participation is the basic condition for the prosperity and development of the sharing economy and is the basic requirement for the realisation of collaborative governance. Encouraging the public to actively participate in the collaborative supervision of my country's sharing economy has become an inevitable requirement to improve governance efficiency and create a social governance pattern of co-construction, co-governance and sharing. First of all, this paper refers to the rapid development of my country's sharing economy in recent years, which has also brought many negative problems. Secondly, this paper establishes a game model of sharing economy-related enterprises, regional governments and the public to drive the public's shared economy in my country, which forms the driving principle and guiding strategy of collaborative supervision. Through research, it is concluded that the size of the initial probability is proportional to the convergence speed of the game subject's selection strategy; different game subjects have different sensitivities to parameter changes, and the public is more sensitive to the cost and punishment of participating in innovation involving different game subjects The strategy choices between them are mutually reinforcing, and the evolutionary stable strategy is ‘active participation, supervision, active participation. Finally, the simulation experiments are used to verify the results of the tripartite game model constructed in this paper.

Entrepreneurship education is an educational concept and teaching mode which is formed in socio-economic development. Universal entrepreneurship education has both the long-term value of realizing the transformation and upgrading of industrial structure and the practical value of alleviating the employment problem of college students. Therefore, how establishing the innovation and entrepreneurship education model in higher education under the environment of big data analysis is a major issue at present and also a key element to achieving the construction goal of modern higher education. In this paper, the entrepreneurship data of colleges and universities are deeply mined by association rules of the Apriori algorithm and clustering analysis of the FCM algorithm. As a result, an entrepreneurship education model is established and practiced in university classrooms. According to the results of the survey after the practice of the entrepreneurship education model, it can be seen that the number of students’ participation in innovation and entrepreneurship lectures has increased, and the proportion reached 50%. Innovation and entrepreneurship clubs formed on campus and discipline competitions accounted for 16.00% and 18.00%, respectively. College students’ motivations for learning innovation and entrepreneurship education courses are mainly ability enhancement, interest-driven, credit demand, future planning, and influence of others, and they account for 34.00%, 26.00%, 21.00%, 14.00%, and 5.00%, in that order. These indicate that the current group of college students considers the pursuit of knowledge and the improvement of their abilities as the main reasons for undertaking entrepreneurial learning activities. After practicing the entrepreneurship education model, people have increased their knowledge of entrepreneurship theory and are more willing to participate in various entrepreneurial activities.

As teachers struggle with the usage of high-quality digital educational resources, small coverage and weak pertinence characteristics of traditional evaluation methods are the reasons for this. In this paper, three aspects including the connotation, operation mechanism and evaluation of ‘Artisan Workshop’ are analysed, and the classification factors are explored based on the collected educational data. By using these objective characteristics, the application ability of teachers to use digital educational resources is quantified, and the prediction model of sustainable development in ‘Artisan Workshop’ is constructed by adopting random forest regression, extreme gradient boost regression and light gradient boost regression models. According to the application of educational resources and teachers’ ability, this paper proposes an optimized scheme for the sustainable development of ‘Artisan Workshop’, which provides a reference for improving the quality of talent training in China.

Paper-cut elements in the design of boat space can effectively improve the homogenisation tendency of existing boat space design, effectively enhance the characteristics of traditional Chinese culture and enrich the development ideas of boat space design. Therefore, based on the Chinese folk art of paper cutting, this paper combines theoretical analysis and empirical research to deeply analyse the artistic expression of paper-cut elements in the spatial design of boats. This paper mainly studies the artistic integration of paper-cut elements in the decorative design of boats and the artistic integration of paper-cut elements in the practical design of boats. Through the calculation and analysis of the sparse autoencoder algorithm model, it can be seen that the decorative integration of paper-cut elements in wallpaper is 82%, which is higher than that of traditional materials (24%). The integration of paper-cut elements in the space design of cabins and guest rooms is as high as 97% and 94%, respectively, which is far better than the integration of traditional materials in the space design of cabins and guest rooms. This shows that the application of folk paper-cut elements to the space design of boats not only highlights the theme of boats but also enhances the cultural value of boats.

Tea culture is the main component of Chinese traditional culture, and the analysis of tea culture dissemination paths can promote the process of Chinese traditional culture dissemination to the outside world. This paper standardizes the tea culture dissemination paths based on the principal component analysis method. The correlation matrix of the standardized data is tested for sampling suitability, and the eigenvalues and eigenvectors are calculated to derive the principal components. The variance contribution rate and the cumulative contribution rate of the variance of the principal components are calculated, and then the scores of each principal component are derived and evaluated comprehensively. Accordingly, the main communication paths of tea culture are new media communication, museum collection and exhibition, and tea trade. Based on this, this paper analyzes the communication effects of the communication paths, and the results show that: the number of followers of public accounts related to tea culture reached 63,214 in 2021, an increase of nearly 24% compared with 2019. The total number of visitors to the museum collection and exhibition of tea culture was 28,004 in 2021, an increase of 22.7% compared with the previous year. The number of tea exports and export countries both increased significantly in 2021 compared with 2012. It can be seen that the main dissemination paths of tea culture obtained by the principal component analysis method are effective for the dissemination of tea culture and also provide a reference meaning for the dissemination of other traditional Chinese culture.

Against the social background of the gradual transformation of the social pension mode and the arrival of the ageing age, the tool design of fitness walkers for the elderly was studied. In this paper, the physiological and psychological status of the elderly with mild or moderate stroke or unable to exercise due to the decline of physical function was analysed. In addition, the corresponding elderly fitness walker product market research was studied. Aiming at these problems, this paper proposes a set of elderly fitness walker frames based on the ergonomic semiotic approach of the product architecture design (SAPAD) model. First, the development and structure of artificial engineering is introduced in detail. Then the framework of walking aid for the elderly based on the SAPAD model is proposed, and the SAPAS model and walking aid for the elderly are introduced, respectively, along with the framework of simulation design and application. The experimental results show that the framework of elderly fitness walker based on the ergonomic SAPAD model can be implemented. It can be applied to the elderly’s fitness assistance. A fitness walker can be suitable for the elderly who want to continue to exercise; it also hopes to care for the elderly who need to exercise from the perspective of the product.

This paper addresses the optimal decisions and channel coordination issues in a green supply chain composed of a socially responsible manufacturer and a fair-mined retailer, where the manufacturer invests in advanced facilities/technologies to improve green quality of products, and the retailer exerts marketing effort to enhance market demand. We develop supply chain models under three scenarios: centralized system, wholesale price (WP) contract without fairness concerns, and WP contract with fairness concerns. Our results show that the retailer’s fairness behavior further causes a benefit for herself, while the manufacturer and the total supply chain to suffer. Moreover, a revenue-cost-sharing (RCS) contract is introduced to coordinate supply chain. We prove that a win-win outcome is reachable, and the RCS contract is applicable in practice.

Customer portrait is the customer information that is labelled and digitised. Collecting and studying consumer data is highly helpful to enterprises because it empowers them to use these data to locate target groups and meet various customer needs. With the continuous rise of the e-commerce market, major e-commerce websites list providing customer value and being data-driven as company values. Traditional customer portraits rely excessively on objective and past experience; so, a simple but powerful customer portrait system is required to make the customer analysis platform simple and intelligent, and to make the data accumulated at ordinary times bring practical value to e-commerce companies. Based mainly on addressing this requirement, this paper proposes an effective customer portrait system for e-commerce. First, the research background and significance is expounded, which lays a theoretical foundation for this research. Then, the customer portrait system is designed, including the overall design scheme and functional framework of the system. The improved FCM algorithm is used as the basis for building the customer portrait model, and the performance of this algorithm is analysed and compared with that of the improved algorithm used in this paper.

In order to give full play to the application of big data in film and television media and imaging in the cloud era, this study proposes a communication-efficient distributed deep neural network training method based on the DANE algorithm framework. The DANE algorithm is an approximate Newtonian method that has been widely used in communication-efficient distributed machine learning. It has the advantages of fast convergence and no need to calculate the inverse of the Hessian matrix, which can significantly reduce the communication and computational overhead in high-dimensional situations. In order to further improve the computational efficiency, it is necessary to study how to speed up the local optimization of DANE. It is a feasible method to choose to use the most popular adaptive gradient optimization algorithm Adam to replace the commonly used stochastic gradient descent method to solve the local single-machine suboptimization problem of DANE. Experiments show that Adam-based optimization can converge significantly faster than the original SGD-based implementation with little sacrifice in model generalization performance. With the increase of sampling rate, DANE-Adam significantly outperforms the DANE method in terms of convergence speed, and at the same time, the accuracy can be kept almost unchanged, which are 0.96, 0.88 and 0.75, respectively. This shows that Adam-based optimization can converge significantly faster than the original SGD-based implementation with little sacrifice in model generalization performance, with significant potential value.

Under the influence of the new energy strategy, the new energy vehicle has developed vigorously. As an important supporting infrastructure, the rationality of its location construction has become the key factor for the popularization of the cars. In this paper, the influence of economy, environmental protection and convenience is comprehensively considered in the model. The comprehensive optimization objective is to minimize the annual cost of construction and operation of the stations, minimize the additional carbon emissions caused by charging, and maximize the service capacity. Considering the layout and capacity constraints of charging facilities in charging stations, the AHP is used to weight each objective function. Thus, the multi-objective optimization model can be simplified, and the chicken swarm algorithm is used to settlement this question.

With the rapid development of computer science and technology, there is an increasing diversity observed in the use of electronic computers. Users browse interactive content such as text, images, audio, video, etc. The increase of the interactive interface results in a slow interface response and affects the user experience. Therefore, this paper mainly studies the user interface under the multi-dimensional optimisation of the Rhino/GH platform, and introduces the long short-term memory and gated recurrent unit algorithms in the visualisation part for optimisation; the study results suggest that the overall response time is 50% but lower than the traditional interface, and the time fluctuation is within 23.7%, which is 23.6% but lower than the traditional 47.3%. When interacting with multiple interfaces, the interaction interface optimised by the Rhino/GH platform maintains a fluctuation range within 29.2%, and the time increases by 13 ms, showing excellent stability and efficiency.

Quality assessment of university education and teaching is an essential and important part of teaching and learning activities. Over the years, a large amount of data has been accumulated in teaching and management, but the potential value of these data has not been fully utilized. How to use a large amount of data scientifically, automatically, and more effectively for rational analysis and guidance of current teachers’ teaching work has become a key issue in testing the effectiveness of teaching quality assessment, and an in-depth study is urgently needed. In this paper, we use the Eviews model, regression analysis algorithm, Apriori association rule algorithm, and other techniques. It conducts research on the assessment of teaching quality in Chinese universities. Finally, after completing the construction of the system, the main modules in the system, system management, resource management, quality assessment, and assessment query module, focused on testing. A total of 160 test cases were designed, of which 157 test results met the requirements of the system, and 3 results did not meet the requirements, with a success rate of more than 98%. The results show that the system has no major defects, runs normally, has stable performance, and meets the system requirements for teaching quality assessment in universities.

BIM technology is a breakthrough in the construction industry whose application is of great value, and the engineering design, engineering budget and cost control combined with BIM technology have been developed rapidly. Therefore, in this paper, BIM technology is adopted to optimise the construction mode, and the genetic algorithm is selected to solve the scheduling model of construction. By applying an interdisciplinary research method combined with computer science and technology, an intelligent optimisation model of construction can provide a new research perspective and technical method for the development of the construction industry.

The use of industrial robots based on MATLAB simulation for water environment monitoring is to monitor the water environment better, improve monitoring efficiency and reduce monitoring costs. The robot can better collect data and can engage in deeper water-specific information. In this paper, based on the discussion of the water environment monitoring robots used in countries around the world for water environment monitoring, we introduce a MATLAB-based simulation of industrial robots in a wide range of water environments to simulate the autonomous data acquisition system. The main advantages are: compared with other robots, it can realize the “wide range” of water environment data collection; compared with fixed buoys, it can realize the “autonomous” collection of water environment monitoring data and gives the autonomous collection process and hierarchical software progression. The autonomous acquisition process and hierarchical software architecture are presented. The simulation results analysis shows no difference between the simulated data and the predicted data from the historical data using MATLAB-based industrial robots for water environment monitoring. This shows that the development of industrial robot simulation in water environment monitoring is promising and feasible.

Art and design are creative activities humans engage in to achieve certain purposes. The multiple characteristics of art and design also put forward more requirements for developing higher education art and design education. In this paper, starting from the problem that the multiple characteristics of art and design make it difficult to evaluate the teaching quality, the FAHP-FCE evaluation system is proposed to score the quality of art and design talents training. Firstly, fuzzy hierarchical analysis (FAHP) is used to construct a judgment matrix to filter out evaluation indexes. Secondly, multi-level fuzzy comprehensive evaluation (FCE) is used to evaluate the evaluation indexes to obtain the score. Finally, conclusions were drawn by comparing the effectiveness of the two teaching models. The research results show that the effectiveness of the FAHP-FCE evaluation system-based art and design talent training teaching model is assessed as 83.47 points, which is 34.11% more effective, 23% more design ability, 31% more learning ability and 26% more hands-on ability than the traditional teaching model. The talent cultivation model in this paper enables students to master abstract design ability, helps to expand the field of creative thinking, enriches the language of modeling and artistic imagination, and improves comprehensive design ability and learning ability.

Ceramic packaging design is the crystallisation of industrial design, ceramic art design and graphic design. Rhino software is used to design ceramic bottles to express the designer’s ideas accurately. Rhino is well suited to a market where new product development cycles continue to shrink. In order to solve the problem of lack of realistic feeling in the teaching of ceramic packaging design, Rhino is used in ceramic packaging design to realise the application of college teaching. In view of these problems, this paper proposes a complete set of Rhino applications in ceramic packaging design teaching. First, a ceramic packaging design process based on Rhino is proposed. PythonScript module supported by Rhino software is used to realise ceramic packaging design through program algorithm. Then, Rhino is employed as a tool for ceramic packaging design, which is very suitable for graphic design students because of its inherent characteristics. Rhino ceramic packaging design will be applied to teaching, making the course construction closely adhere to the local industry, which is conducive to expand the employment opportunities available to students. The experimental results show that Rhino ceramic packaging design can be applied to teaching, and this capability indicates that it is possible to endow future teaching activities with a basic innovation research direction.

In order to improve the recognition accuracy of symphonic music contour, this paper constructs an intelligent music main melody recognition system based on artificial intelligence technology to make melody recognition with certain search adaptation capabilities. Based on the traditional melody recognition system, the fundamental tone sequence of symphony fragments is obtained by using the fundamental tone extraction and short-time autocorrelation function in the melody contour feature extraction algorithm, which is transformed into the melody contour sequence after regularization and merging to determine the similarity of the music melody signal itself. The wavelet transform method and radial basis function algorithm are used to improve the defects of monophonic discrimination in the traditional recognition model so that the artificial intelligence technique can effectively fit with the symphony recognition model of music melody contour. The experiments show that: The average recognition accuracy of the AI-based music melody recognition system is 90.5%, which is significantly better than 69.5% of Sound Hunter software and 76.5% of Shazam software. For the five monophonic chords, the system’s recognition accuracy is as high as 98.3%, especially in the field of hanging chords with significant recognition effects. It can be seen that the artificial intelligence-based music main melody recognition system provides a scientific and authoritative recognition means for the dissemination and development of symphonic music and is conducive to improving the recognition accuracy of symphonic melodies.

Constantly promoting theoretical innovation is the key to the perpetual vitality and vigor of Marxism. Marxist theory interpretation can provide new perspectives, new assertions, and new requirements for adhering to the basic principles of Marxism in line with the requirements of the times, national characteristics, and practical needs, and is the main basis for the formation, development, and renewal of the theoretical paradigm. Therefore, Spring+SpringMVC+Mybatis is used as the back-end development framework of the decentralized Internet to establish the decentralized Internet model used for interpreting Marxist theories. It is realized through the stages of theoretical semantic representation, multi-dimensional feature extraction, and classification. According to the multidimensional interpretation effect, the classification recall of the theoretical interpretation dimensions obtained after using the decentralized Internet model increases from 59.15%, 58.84%, 61.21%, 62.79% to 69.49%, 72.03%, 71.87%, 72.9%, and the average running speed of 9.21s decreases significantly to 3.84s. Portability, accessibility, interpretation completeness and contextual integration, depth of understanding indexes of ranking, accessibility, interpretive completeness, and contextual integration, depth of understanding, interpretive adequacy, and six-theory coverage reached 89.26%, 91.45%, 90.75%, 92.84%, 93.23%, 96.29%, and 99.12%. It shows that the decentralized Internet interpretation method of Marxist theory can grasp the scientific connotation and inner mechanism of theoretical development, trace the origin from ideas and perspectives related to the concept and summarize and interpret the background, reality, gist, content structure, and its contemporary significance under the new era discourse, so as to better implement and practice Marxist theory.

In order to explore the development path of Hunan embroidery under the vision of artificial intelligence, promote the digital regeneration and database construction of Hunan embroidery stitches, the communication, and interaction between Hunan embroidery brands and the public, and create more possibilities for revitalizing the culture and industrial development of non-heritage Hunan embroidery. In this paper, a mechanics model of Hunan embroidery stitch is established based on the finite element idea under the view of artificial intelligence. The single yarn in the yarn is regarded as a frictionless articulation of some rows of elastic rods with a circular cross-section. The elastic rod can only be subjected to axial force without a moment, and it is a uniform, continuous, and completely elastic isotropic body. Using the displacement method, the displacement of the unit node is taken as the basic unknown quantity, the displacement in the unit is assumed to be linearly distributed, and the displacement of any node in the unit is obtained by linear interpolation. The strain, stress, and stiffness matrices of the elastic rod unit are derived, the equilibrium equations are given, and a database is established. The results of the study showed that consumers of all age levels thought that the patterns representing Hunan embroidery mainly include Hunan characteristic landscapes, portraits of Hunan great men, traditional flowers, birds and animals, and totems of Chu culture, etc., among which Hunan characteristic landscape accounted for 54% of the largest proportion. It provides a development direction for the inheritance and protection of Hunan embroidery skills.

The teaching mode of the innovative automotive manufacturing process is to improve students’ automotive manufacturing technology. In this paper, firstly, the basic principle of the ant colony algorithm is introduced, and the ant colony algorithm is improved and optimized by heuristic function, state transfer probability and pheromone update rule. Then the optimized ant colony algorithm is used as the base algorithm for constructing the ADDIE model. After establishing the ADDIE model, two dimensional indicators of the teaching mode of the automotive manufacturing process, i.e. teaching objectives of the basic course and application course, are mined for the example of the automotive manufacturing process course of L school. The two example indicators proposed are analyzed and illustrated using the example data.

Regarding the basic course teaching objectives, the mean values of A, B, C and D evaluation indicators accounted for 47.73%, 21.46%, 20.17% and 10.65%, respectively. Regarding the teaching objectives of the application courses, the mean values of the four evaluation indexes of A, B, C and D account for 50.24%, 15.75%, 21.05% and 12.96%, respectively. The innovative teaching mode of the automotive manufacturing process based on the ADDIE model is beneficial for students to learn technology in a new way and also for delivering high-quality technical talents to automotive manufacturing enterprises.

Wear of seal materials is a widespread problem in the aerospace, petrochemical, and electric power fields, leading not only to low efficiency of mechanical equipment, but also to increased energy consumption and reduced safety performance. Therefore, this paper establishes the SCA-wear model based on the tensile constraint algorithm to calculate and analyze the friction and wear performance of mechanical seal materials. The friction coefficients of different loads at 200°C and 10 min after the test time stabilize, and fluctuate in the range of 0.35 at 30N, 0.26 at 70N, and 0.48 at 90N. The frictional wear of the C/C composite mechanical seal material is at least 15.6%. The reason is that it is composed entirely of carbon. It has many advantages of carbon and graphite materials. Therefore, the frictional wear of the C/C composite seal material is low.

Researching the past empirical studies, we find that the empirical results of using data from the same or similar stages of economic development tend to be very close; however, there will be some discrepancies when comparing the empirical results of data samples at different stages of development. In order to explain this phenomenon, a theoretical analysis is made on the regional differences in the impact of population ageing on national saving. By analysing the areas considered in this paper as being in different stages of economic development, the combined effects of population ageing on the national saving rate are different. In order to verify the above conclusions, the thresholds and threshold effects are estimated and tested through the threshold model. The results show that the impact of population ageing on the savings rate will be different due to different levels of economic development. When per capita income is below the threshold of 9001.69, population ageing has a greater negative impact on the national savings rate. When per capita income is above the threshold of 9001.69, the negative impact of population ageing on the national saving rate is smaller.

The economic benefit of enterprises is an effective index to measure the economic activities, which forms the basis and starting point. Therefore, in this paper, the traditional analysis methods of enterprise in terms of economic benefits are compared and based on its evaluation principles, the advantages, theory and realisation process of the kernel function in enterprise economic benefits analysis are analysed. Then, different kernel functions are selected to analyse the total output value, product sales rate, sales revenue, total profit and total profits after taxes of five enterprises, and their economic benefits are evaluated by the cumulative contribution rate of principal components. Finally, the mixed kernel function based on the combination of polynomial kernel function and Gaussian kernel function is used to optimise the analysis effect, which is meant to help enterprise leaders make more scientific decisions and lay a foundation for the sound development of enterprises.

In order to have a better product display and thus attract consumers’ purchases and increase the economic benefits of the enterprise, in this paper, we propose a deep learning model for brand 3D image design. A feedforward neural network that estimates the error of previous layers based on the error of the output layer assigns the convolutional kernel weight parameters of the network in the interval and stops when the error reaches a preset accuracy or reaches a preset maximum learning count. The locally-aware convolutional neural network acquires local features that are finer than the global features and outputs the feature maps of the convolutional layers after passing the activation function to calculate the sensitivity of the sampled layer units. Given the sensitivity information of the feature map, the gradient of the kernel function weights is obtained, and the updated parameters are trained to achieve feature map recursion and solve the image boundary problem. A 3D recurrent neural network is constructed using data-driven multiple or single images, transformed into a low-dimensional feature matrix, processed with 3D pixel data, extracted perceptual features, and generated high-resolution images. The analysis of the results shows that the CD value of the used model is 0.477 and the EMD value is 0.579, which makes the constructed 3D images with more obvious detail levels and more accurate structural design, while the model of Pixel2Mesh focuses more on surface information, so the generated model is more realistic and closer to the real image.

Digital humanistic knowledge production emphasises the importance of a strong knowledge production community and differentiates from traditional knowledge production models, which include aspects such as online and cooperative knowledge development. The digital humanities knowledge production community model is already widely acknowledged. However, the features and characteristics of digital humanistic knowledge production under natural language processing are controversial. This research presents a wordVEA digital humanistic knowledge production feature mining approach based on a word2vec and variational self-encoder (VAE). The knowledge production characteristics of digital humanistic are primarily defined by the coexistence of a knowledge production structure and boundary blurring, as well as interdisciplinary collaboration thematic cohesiveness and broad horizon, as determined by the research results which effectively address the question of the characteristics of digital humanistic knowledge production through application of the word VAE method.

With the rapid development of Internet technology, the field of hotel management has begun to carry out informatisation construction, through which the management level and operational efficiency of hotels can be improved. The traditional hotel management work with a complex structure and cumbersome process is facing huge challenges. Therefore, this paper aims to design, develop and implement a small and medium-sized hotel management system based on customer relationship management (CRM), according to the characteristics of small and medium-sized hotel management. Management and system setting management of small and medium-sized hotels can reduce the hotel operating costs and increase the profitability by using this system. The department and employee performance appraisal can be divided into four dimensions, finance, customer, internal business process, and learning and growth for job analysis and design, using the key performance indicator method to lock key appraisal indicators, and ignore or omit irrelevant or insignificant indicators. Then the analytic hierarchy process is used to scientifically determine the indicator weights. The Delphi and in-depth interview methods are used to complete construction of the performance appraisal system, and the constructed CRM performance management system is used to achieve the hotel's strategic goals, thus completing the fundamental goal of using the CRM model. The slacks-based measure model is based on the assumption of variable returns to scale, considering that the input redundancy and output deficiency of ineffective decision-making units are measured based on their slack. This paper defines input inefficiency as the ratio of input slack to actual input, and output inefficiency as the ratio of output slack to actual output. The total input inefficiency rate of the hotel industry in China's ineffective provinces and regions is 0.458, while the total output inefficiency rate is 0.077, indicating that if the total investment in the hotel industry is reduced by 47.8% on average and the total output is increased by 6.9% on average, the hotels in these ineffective provinces and regions can all be relatively effective. Finally, the overall performance of the system will be evaluated, and a set of safe, convenient and friendly hotel management system will be developed.

With the rapid development of the market economy, a large number of enterprises provide many jobs with different requirements. In the traditional application process, college students need to search the job requirements of each company one by one to match their own needs and conditions, which not only requires a lot of time and opportunity costs, but also has poor matching degree. This paper uses the recommendation and machine learning algorithms to match and optimise the job characteristics and needs according to the professional type, interest and specialty, employment area and personal preference of college students through the algorithm, and recommends suitable positions for college students to improve their success in application and increase their employment satisfaction rate.

In the construction of intelligent teaching resources, a large number of resources are collected to meet the teaching needs, and its organisational structure, resource content and teaching quality are not fully recognised. At present, there are few studies on the performance evaluation of intelligent teaching, and the research angles are also different. Based on this, this paper constructs the application performance evaluation model of intelligent teaching resource organisation mode. Based on the simple analysis of intelligent teaching research and the development of big data mining algorithm, this paper uses data mining algorithm to realise resource mining, uses ant colony algorithm and association rules to mine information, carries out student evaluation on this basis, and uses ID3 algorithm to construct decision tree to analyse the main factors affecting teaching. Then the performance evaluation model is built, and the concept of fuzzy mathematics is introduced to realise the comprehensive evaluation based on analytic hierarchy process. Finally, simulation tests and experiments are used to analyse the effectiveness of the algorithm and the application of the model. The results show that the data mining algorithm based on association rules can mine more information and shorten the running time. The performance evaluation model based on fuzzy comprehensive evaluation can be applied to the evaluation of intelligent teaching, find the shortcomings in teaching, and confirm the effectiveness of the model.

In the process of designing animation characters and background elements, there may be a high degree of visual blur, resulting in a strong sense of picture distortion, and it is easy to expose the incomplete picture. Accordingly, this paper proposes to use directional derivatives to deal with animation-like visual problems. This method is based on the theoretical basis of directional derivatives, combined with the current mathematical methods of animation character and background element design, and defines the important role of directional derivatives in animation and background design, and more effectively enhances the visual results of animation characters. In the research of clarity and distortion, this paper uses directional derivative derivation as the cutting-in method to test the algorithm for visual simulation and restoration of animated characters and background elements so that the algorithm can be used for each animation character and background element. The defect point is calculated, and the clarity and self-healing ability of the video itself are improved by the influence of the mathematical parameters of the surrounding known points on itself and the key variable of the directional derivative in the field. The results show that the directional derivative can play a role in promoting sublimation in the design of animated characters and background elements.

Studying the concept lattice of intuitionistic fuzzy language can not only enable an intuitive expression of the relationship between objects and attributes but also ensure the effectiveness of information, and it thus provides a new idea for the development of artificial intelligence. Therefore, in this paper, the rule extraction method is studied based on the concept lattice of intuitionistic fuzzy language. By combining the formal context of intuitionistic fuzzy language with the context of decision formal, the formal context of intuitionistic fuzzy language decision is put forward and the rule extraction method under this context is constructed. The problem of decision making in real life can be solved through the weighted similarity in the formal context of intuitionistic fuzzy language.

The manufacturing industry requires a unique recommendation system to suggest products and raw materials, but its performance is often poor in massive data environment. In order to solve the similarity connection problem of large-scale real-time data, the optimised incremental similarity connection method which is used to deal with streaming data can be used to concisely obtain the longest common additive sequence of two given input sequences. This paper, on the basis of the recursion equation, applies a very simple linear space algorithm to solve this problem and adopts new states to carry out similarity connection of incremental data. The experimental results demonstrate that this method can not only ensure the accuracy of real-time recommendation system but also greatly reduce the computed amount.

Today, corresponding to the new wave of internationalisation and integration, education is also subject to rapid and innovative changes. Accordingly, detailed research is undertaken in the present study on innovative teaching strategies that can be developed, under the background of artificial intelligence, to familiarise college students with the ability to fluently express their ideas in English. The research shows that among the factors affecting the quality of English teaching, teachers’ teaching methods, English teaching environment and students’ independent learning ability account for high weightage, which are 54%, 31% and 11%, respectively. Through the model of college students’ English teaching innovation strategy, it can be seen that the teaching method, teaching environment and autonomous learning method are the most feasible methods that can be employed in crafting a students’ English teaching innovation strategy.

In recent years, artificial intelligence has gradually become the core driving force of a new round of scientific and technological revolution and industrial transformation, and is exerting a profound impact on all aspects of human life. With the rapid development of Internet big data and high-performance parallel computing, relevant research in computer vision has made significant progress in the past few years, becoming one of the important application branches in the field of artificial intelligence. The exercise of image classification forming part of computer vision tasks involves a large amount of computation, and training based on traditional deep learning (DL) classification models typically involves slow training and low accuracy in many parameters. Thus, in order to solve these problems, an image classification model based on DL and SAE network was proposed. Firstly, the main research of computer vision task-image classification is introduced in detail. Then, the combination framework of deep neural network and SAE network is built. At the same time, the deep neural network was used to carry out convolution operation of the parameters learned by SAE and extract each feature of the image with neurons, so as to improve the training accuracy of the deep neural network. Finally, the traditional deep neural network and SAE network were used for comparative experiment and analysis. Experimental results show that the proposed method has a certain degree of improvement in image classification accuracy compared with traditional deep neural network and SAE network, and the accuracy reaches 97.13%.

In recent years, the investment amount of university engineering construction has been increasing significantly, and the establishment of scientific and reasonable cost management system has become particularly important in the investment control of engineering construction. This paper explores the investment control of university engineering construction and designs the basic framework based on three parts: whole process cost control, full caliber cost audit and full gradient grading management. With the help of BIM, we build a cost management system, which is divided into the planning stage, bidding stage, mid-construction, until the completion of the collation of cost data, extraction and preservation of cost indicators. After establishing a full range of cost management systems, the adjacency matrix X algorithm and functions are used to calculate. The experiment proves that: the investment estimate before the project construction affects more than 75% of the overall investment, and the all-round cost management system built based on BIM can make the investment estimate accurate and stable at 75%, and the negotiation change rule based on this is to save nearly 25% of the investment funds. The comprehensive cost management system has a comprehensive and detailed grasp of the cost process of specific construction projects and specific issues during construction, fully protecting the university’s investment interests.

In recent years, the development of artificial intelligence technology and theory has been rapid, and the application in language science has been gradually comprehensive and diversified, especially the accuracy rate of artificial intelligence for Chinese language is up to 90%. In the era of artificial intelligence, the effect of different structures and parameters of arithmetic models on Chinese language recognition varies greatly. Language science is an important research area for realizing machine-human communication, and accurate comprehension of the meaning of linguistic expressions is the key to realize communication. In this paper, we construct a speech system that is different from the traditional stable time series for the irreplaceable characteristics of artificial intelligence technology to improve Chinese language ability. A dynamic Bayesian network (DBN) is used for modeling and analysis, and a DBN construction method is investigated to import a hidden Markov model in a speech recognition system to reveal the interactions between nodes within multiple time slices. The accuracy of dynamic Bayesian networks in Chinese dialect inference algorithms is demonstrated using Matlab simulations to characterize the reliability of speech features using a speech spectrogram. It proves that artificial intelligence technology and Chinese language science are complementary and mutually reinforcing, showing a good and rapid development trend.

In order to improve the ability of micromechanical sensors to update data-matching nodes in real-time and speed up the establishment of a neighbor relationship between nodes, this paper proposes a personalized recommendation algorithm for micromechanical sensors based on the cloud model to improve the stability and real-time performance of micromechanical sensors. The algorithm uses the service attribute values of the cloud computing model and cloud clustering method to set feature term labels and establish a cloud service similarity matrix so as to meet the user-matched manufacturing service requirements. The intra-class clustering technique is applied to measure the clustering effect, and the evaluation function of each clustering number is calculated on the basis of considering the time sequence to determine the best clustering result to complete the personalized recommendation path for micromechanical sensors. To verify the application effect of the personalized recommendation algorithm for micromechanical sensors based on the cloud model, experiments are conducted. The results show that the recommendation algorithm in this paper can always control the node energy consumption below 2.5×103, and the average discovery delay is stable between 41-43 seconds. And the sensor response time is 12.1 seconds, and the average absolute deviation value is 0.23, which is nearly 1.3 times smaller than 0.53 and 0.52 of the collaborative recommendation algorithm and hybrid recommendation algorithm. It can be seen that the recommendation algorithm in this paper solves the problem of excessive neighbor discovery delay in the communication process of micromechanical sensors and effectively improves the personalized recommendation performance of micromechanical sensors.

In order to reduce the vibration damage of the engine and improve the automatic transmission adjustment performance of the automobile, an automatic transmission system based on torsional damping is constructed in this paper to optimize the hardware function of the shift speed. The torsional damper is used as the core control element to reciprocate the automotive mechanical automatic transmission system, and the calculation method of torsional vibration analysis of the crankshaft shaft system is used to determine the indexes of rotational inertia, stiffness, and damping parameters of torsional vibration. The damping coefficient values are set near the equilibrium point, and the loss factor of the automatic transmission system is signal processed to obtain the best damping effect. Using the intrinsic frequency as the central processing unit of the automatic transmission system, the differential equation is used to calculate the excitation torque vector of each cylinder of the engine, and the characteristic module of the automatic transmission system is developed. The tests showed that the cylinder burst pressure of the automatic transmission system with the torsional damper reached a maximum of 67.9 pressure at 4800 amp. The crankshaft front-end torsional angle of the 4th-order main harmonic excitation reached a maximum of 0.57 degrees, which exceeded the engineering allowance of 0.2 degrees and reduced the automatic shift shock by about 28.36%. It indicates that the torsional damper can improve the control performance of the automatic transmission system, which is conducive to enhancing the stability and self-adaptive capability of the transmission adjustment.

The computer vision direction in the field of artificial intelligence analyses the latest progress of computer vision technology from visual perception and visual generation, including but not limited to image recognition, target detection and image segmentation. First of all, for computer vision technology, this paper introduces the detailed application of image recognition technology, object detection technology and image segmentation technology. Then, we build a BP neural network combined with a deep LSTM neural network, use the BP network algorithm to select the input variables to reduce the dimension and complexity of the model, and use the selected variables as the input of the deep LSTM network. At the same time, deep LSTM is used to perform high-dimensional deep memory learning features on the selected variables. Finally, the model is separately experimented in computer vision. The experimental results show that the present model and other single models can be selected by BP neural network variables in computer vision applications, which can effectively reduce the complexity of the model and improve the generalisation ability of the model, so that it can be used in computer vision research.

The analysis of the psychological impact of the spread of Yangming studies in Japan on the Japanese people is to enable Yangming studies to be better developed in Japan. Based on big data analysis technology, this paper constructs a hybrid data analysis model using the EM algorithm and proposes performance evaluation indexes for the model. Under the EM data analysis model constructed in this paper, the example indicators of the Japanese people’s psychological impact in disseminating Yangming studies by big data analysis are explored, i.e., the psychological acceptability of the dissemination method and the psychological and moral construction impact. Regarding the dissemination method, the Japanese people are more receptive to disseminating Yangming studies in Japan through “learning rules”, with an average percentage of 39.37%. Regarding psychological and moral construction, 90.22% of the Japanese people believe that disseminating Yangming studies can promote self-improvement of value standards and correct self-examination. Based on the big data analysis, we can effectively see from the data the impact of Yangming studies on the audience in the process of dissemination, and improve the scope of Yangming studies dissemination according to the data feedback, so that more people can recognize the idea of unity of knowledge and action.

Because of its large structure and strong bearing capacity, large rolling bearings are mainly used on low-speed and heavy-duty occasions. At present, the analysis and Research on large rolling bearings include load distribution, bearing capacity, bearing service life and bearing structure optimisation. Taking the large four-point contact bearing as an example, this performance studies the method of channel structure parameter design and life analysis. First, the structure and simplified model of four-point contact ball bearings are designed, and then the life analysis and calculation model is defined through analysis. Through the analysis of bearing finite element calculation results, the parameter optimisation of bearing channel spacing and channel curvature radius coefficient is studied, and the parameter design is carried out. Finally, the fatigue life analysis shows that the calculation result of the stress life model is 1.699 times that of strain life. The research results have guiding significance for the design of large four-point contact ball bearing and its supporting structure.

This study uses the event study method to study and analyse the impact of the release of policy information related to the COVID-19 epidemic on the changes in the stock prices of listed companies of property service enterprises in China. The results show that the Hong Kong capital market has been greatly affected by the release of policy information related to the COVID-19 epidemic. Additionally, the study demonstrates that the policy effect of the introduction of policy information related to the COVID-19 epidemic exists in the short term, and that the effectiveness of policy information related to the COVID-19 epidemic will change over time.

Against the background of the current COVID-19 pandemic and the popularisation of the Internet, the demand for online teaching is increasing in colleges and universities. But a course in music like piano teaching that requires multi-sensory learning still presents great challenges. Given this background, this paper analyses the system functional requirements of online piano teaching in colleges and universities, designs the overall system architecture including software and hardware, selects the wireless network communication method after analysis and judgement, and designs a remote wireless network-based system. Finally, the advantages and disadvantages of different protocol algorithms are compared.

In order to make spoken English more widely used in college communication, the author proposes the application of ARCS motivation model in college spoken English teaching. The author tries to apply ARCS motivation model in the strategy design of college oral English teaching, the four core factors of attention, relevance, confidence and satisfaction are explained, analyzed and explained in detail respectively, twelve practical teaching methods and strategies are summarized. A total of 109 students in 3 classes were taught for a semester, the teaching strategies and methods mentioned above were used in the weekly oral English teaching classes. After applying several teaching strategies designed by ARCS model to stimulate motivation, a post-test was conducted for all the students in the experimental class, by comparing the data before and after, the anterior and posterior sides of S1 natural consequences completely disagree: 14.67%, 13.75%, less agree: 29.35%, 24.76%, unsure: 11.04%, 1.82%, more agree: 19.26%, 33.07%, strongly agree: 25.68%, 26.6%. We can see that in terms of attention, relevance, confidence and satisfaction, the results show that ARCS motivation model is effective in college oral English teaching.

The Internet of Things, as an important part of important data aggregation, forwarding and control, often leads to objectivity errors due to the huge and complex received data. Based on this, this paper introduces GRU, LSTM, SRU deep learning to optimize the data received by the Internet of Things, and selects the most suitable communication mode optimization algorithm. The experimental results show that the accuracy errors of GRU, LSTM, and SRU algorithms show a downward trend, from 0.024 to 0.010%; the training time is reduced by 254 minutes, and the training speed is increased to 86%, indicating the excellent performance of SRU deep learning in IoT gateways.

With the advent of the artificial intelligence era, the traditional methods of online learning can no longer meet people's personalised needs. Take the English vocabulary memory APP, which is widely used in the market of learning software, as an example. At present, there are various factors that make it difficult for learners to improve their learning efficiency, reduce their learning enthusiasm and enhance their learning effects. Therefore, in this paper, the application of machine learning algorithm in intelligent recommendation of English vocabulary learning is studied, and an intelligent recommendation system for English vocabulary learning based on crowdsensing is established. In its functional module, machine learning algorithms such as clustering algorithm, word vector training and collaborative filtering recommendation are used to realise the intelligent recommendation in the system. Finally, accurate resources of recommendation for user groups with different needs, interest characteristics and abilities improves the efficiency of English vocabulary learning.

This paper mainly discusses the formation and development of the core literacy of sports majors. Then, this paper establishes the factor model that affects the professional quality of youth sports employing multiple linear regression analysis. First, this paper uses factor analysis to observe 19-factor patterns. After the maximum variance rotation method, four dimensions were determined from the factor loading matrix, and confirmatory factor analysis was performed. Finally, this paper uses the multiple linear regression method to establish young students’ core literacy education model. This paper analyzes the model and draws the influence of four factors of school, family, society and individual on adolescents.

The paper takes the data of a 50 MW photovoltaic power generation system as a sample, divides the weather conditions into two categories according to whether there is a sudden change, optimises the decomposition number K and penalty factor of variational mode decomposition (VMD) by using the sparrow intelligent algorithm, decomposes the power sequence in a power mode by using the optimised VMD decomposition method and sends all sub-components to a long short-term memory (LSTM) network for prediction.

In order to have a qualitative improvement in the education management of universities in the background of the era of big data technology, it is necessary to integrate information technology and technological means in the process of education management. Based on this background, this paper constructs an informatization education management platform. In the process of platform architecture, the cloud storage module obtains the exact nearest neighbors by the time factor, optimizes the construction process by using the weighting function and hybrid coordination filtering algorithm, and simplifies the teaching management process. In the teaching management system, the fixed rules of the feature selection algorithm are used to achieve teaching resource selection prediction in the optimal feature subset and establish the best classification surface. Finally, the performance of the platform proposed in this paper is verified from two levels: resource recommendation and education management. The experimental results show that the MAE (Mean Absolute Error) value of this platform is always under 0.7, with a mean value of 0.523, and the experimental teaching resource recommendation effect is better than that of the IE virtual experiment teaching platform and data platform. And the maximum number of visits that the platform can bear is 884, and the maximum amount of government processing can reach 839, which is much larger than the number of 800 users of experimental teaching in the university and fully meets the needs of university education management.

Internet information capacity is expanding and developing at a high speed every day. The huge amount of information data involves many problems, such as resources could be scattered, overlapped and confused and thus difficult for users to timely and accurately capture information suitable for them, especially in the field of education. Through the MRLG Rec algorithm, the following processes can be carried out: deep processing, characterisation of data to generate unique individual portraits of user information and to store education resource databases at the same time and analysis of recommendation system to provide users with more accurate high-quality education resources, thus forming a two-way unblocked and efficient information transmission closed-loop, which help learners find teaching resources of the time. It also has screening channels of high-quality educational resources. Through the analysis of experimental data, it is concluded that the algorithm improves the user viscosity of human-computer interaction in the teaching resource recommendation system.

Wushu Sanda is a practical fighting skill and a new modern competitive event. Under certain conditions, they use kicking, hitting and wrestling as the main means according to certain rules. A good Sanda athlete should have good physical quality and special skills. Given the continuous change and revision of Sanda rules, the means to reduce the number of injuries sustained by athletes, tap their potential and prolong their sports life has become an urgent priority. During the training process, the coaches should use a guided and inspiring educational model to enable athletes to think positively and rationally. It is necessary to clearly understand the physical conditions of the body, combine the special characteristics of the Sanda movement and adopt effective prevention and recovery methods. This paper takes the sports fatigue of Sanda athletes as the research goal. To this end, we use physical, biochemical and psychological indicators and means to comprehensively monitor athletes’ sports fatigue. In order to improve martial arts training, improve the physical health of martial arts athletes and improve the athletes’ competitive level.

To identify the influence of the lack of audit information on the decision-making of securities investment, a method to analyse the impact of missing audit information on the decision-making of securities investment based on two Logistic regression models was proposed in this paper. Assumptions about the impact of missing audit information on decisionmaking of investment were developed. Two Logistic regression models were established to analyse the impact by using Logistic regression functions. Variables in analysis models were designed to solve the Logistic regression models with an ADMM algorithm, so as to analyse the impact of missing audit information on the decision-making of securities investment. The analysis of the impact of missing audit information on the securities investment decisions based on two selected models yielded the following results: (1) The lack of audit information was positively correlated with other decisions except the standard effective securities investment decisions; (2) To improve the problem of audit information missing, as the time went on, there was a strong correlation between the lack of audit information and the effectiveness of securities investment decisions; and (3) The lack of audit information was negatively correlated with the effectiveness of securities investment decisions.

This paper aims to construct a multimodal poetry translation corpus for easy retrieval by poetry translation researchers and enthusiasts. In this paper, we combine the AdaBoost model and ELM network and propose the ELM-AdaBoost method to put the existing poetry translation corpus into the ELM network for learning, to obtain several weak predictors, and then use AdaBoost for classification iteration, and update the weights according to the prediction sequence weights, and then obtain strong predictors, and finally construct a multimodal poetry translation corpus. The search of this corpus shows that, in terms of the ideographic performance of the translations, the ancient poems perform the best, followed by the five-line stanzas, with mean evaluation scores of 83.2 and 80.9, respectively. The seven-line stanzas have the best phonetic performance, with an average rating of 73.2. The rhetoric of five-verse poems was the best, with an average rating of 63.5 marks. The overall translation effect is relatively poor because the meaning is often difficult to account for, or there is a cultural gap in the translation of poems. The multimodal translation corpus based on the AdaBoost model is a powerful tool for poetry translation research, which provides strong data support for Chinese poetry translation research and is of great significance for Chinese poetry culture.

Image inpainting aims to fill the undetectable domain and has been studied using deep learning in recent years. This study investigates smoothness-informed convolutional neural network models for image inpainting. The total variation (TV) is considered and local smoothness constraints are also explored in this study. The local smoothness constraint is conducted by the fidelity of the low-order derivatives on mostly connected parts of the given image at training stage. Unlike most neural network-based inpainting methods using numerous images for training, only a single local image containing the domain to be filled is required for the whole training here. The convolutional neural network accepts the image and is trained using detectable data. Computational results indicate that the local smoothness constraint can conduct a more satisfactory inpainting in comparison to usual TV-based one. We also demonstrate how a deep learning approach is used to solve the Euler-Lagrange equation-based inpainting.

This thesis is based on data obtained from conducting investigations concerning financial literacy among rural households in Shaanxi and Yunnan in 2016. The study chooses the Heckprobit model based on the instrumental variable for conducting its analysis, identifies the influencing factors of rural household credit demand, especially the credit constraint, and conducts high-spot reviews on the effect and heterogeneity of financial literacy. The research shows that the financial literacy has an obvious positive impact on the credit demand, i.e. the higher the financial literacy level of the rural household, the bigger the credit demand that the rural household will have; the financial literacy has an obvious negative effect on the formal credit constraint, without any prominent impact on the informal credit constraint or the overall credit constraint. It means that promotion of the rural household’s financial literacy level can only ease the formal credit constraint, and worse, the higher the financial literacy level is, the weaker the formal credit constraint of the rural household will be.

For a long time, the situation of students’ learning in physical education (PE) was not optimistic, especially the basic movement learning after class, which lacked effective online learning tools. With the in-depth research of deep neural network and the rapid development of computer hardware, the artificial intelligence technology based on deep learning has performed well in the field of basic teaching. Therefore, in this paper, an intelligent teaching system of basic movements in PE is designed. First, the information of coordinate points is collected according to the Gaussian model, and the pose of students is estimated by OpenPose. Second, the overall architecture and functional modules of the system are designed. Finally, the deviation limbs that affect the standard of overall movements are identified by the matching algorithm, which realises the evaluation and feedback of basic movements in PE. Through this teaching system, teachers can obtain the learning situation of students’ movements, and students can adjust their movements through the feedback, which achieves the convenient interaction of PE teaching.

In the context of vocational education reform, with the rapid development of big data and internet technology in recent years, coupled with the increase of employment pressure in China, universities have paid more attention to students’ employability competitiveness and paid attention to the reform and innovation of employment education and management work around the goal of higher vocational education personnel training. Based on the theory of the GRU employment education model and supported by big data internet technology, this paper uses a questionnaire survey and semi-structured interview to conduct exploratory factor analysis and validation factor analysis on the constructed employability evaluation index system by using empirical analysis method and finally constructs an employment education model for higher vocational students. At present, compared with the nearly 100 years of exploration history of employment education in colleges and universities in developed countries such as the United States, the employment education of college students in China has a relatively short history and is still in the exploration and development stage. The combination of vocational education reform and big data and internet technology will create a good environment for the future development and growth of students. Therefore, it is necessary to improve the relevance of employment education and management from the perspective of students and teachers in the “Internet+” environment and provide some new ideas to solve the outstanding problems in the employment education of college students so as to ensure that students can enter and adapt to the workplace smoothly after graduation.