Research on the Application of Machine Learning for Marxist Education Integrating Chinese Excellent Traditional Culture

One of the important reasons why Marxism can stay alive forever is that it can keep pace with the times and integrate with different theoretical doctrines, which is a worldwide and internationalized doctrine [1–2]. As Marxist theory became popular all over the world, China began to introduce Marxist-Leninist ideas and gradually realized the Chineseization of Marxism according to China's national conditions, thus becoming the ideological force supporting China's continuous development and realizing national independence and national unity [3–5]. On the basis of drawing on the excellent cultural achievements of mankind, Marxist theory incorporates the essence of major theories, fully reflecting its theoretical quality of keeping abreast of the times and constantly updating. It is precisely because Marxism's tolerance and openness creates the possibility of its integration with traditional Chinese culture [6–9].

Chinese traditional culture is the valuable spiritual wealth of the Chinese nation, which is the crystallization of the wisdom of the working people and the highly condensed summary of the practical experience of production and life. The protection and inheritance of excellent traditional culture in the new era is an inevitable requirement for promoting the high-quality development of the socialist cultural cause, and it is the manifestation and embodiment of historical self-confidence and cultural self-confidence [10–12]. To promote the development of outstanding traditional culture in the new era, it is necessary to unswervingly adhere to the guiding position of Marxism in the field of ideology, based on the practice of promoting the integration of Marxism and Chinese outstanding traditional culture, and activate the contemporary value and meaning of the outstanding traditional culture in innovative development and creative transformation, so as to revitalize its vitality and vitality [13–16].

Literature [17] points out that the integration of Marxism and Chinese traditional excellent culture is the inevitable result of historical development, Chinese excellent culture promotes the localization of Marxism, and Marxism promotes the innovation of Chinese excellent culture. In the context of the new era, the integration and development of the two play an important role in establishing cultural confidence and socialist modernization. Literature [18] indicated that education has undergone a great change under the impetus of big data, especially the Civic and Political Education in colleges and universities. The challenges faced by this phenomenon and its impact on online Civic Education in colleges and universities are discussed, and solutions are proposed in the context of cases with the aim of promoting the modernization of Civic Education in Chinese colleges and universities. Literature [19] takes traditional Chinese philosophical culture as an object, explores its relationship with the concept of higher education, and analyzes the relationship between the two in terms of interaction and significance. The results of the study specified that the interaction between the two is an important breakthrough for educational reform to show new achievements and has significant meaning. Literature [20] analyzed the guiding position of Marxism in ideology and the position of its ideas in traditional Chinese philosophical culture, and explored the ideas of traditional Chinese philosophical culture to consolidate the guiding position of Marxism. Literature [21] points out that cultural identity is a kind of value norm, which embodies the common value aspirations and expectations of the nation. And Marxism, as a methodology for transforming the world, needs corresponding subjects to undertake this mission in both transformation and understanding. From this point of view, the value norms of Chinese excellent traditional culture is the subject to undertake the cause of Marxism in China. Literature [22] answers the question of localization of Marxism in China by combining the Chinese excellent traditional culture. It discusses the combination of the excellent Chinese culture with the localized culture and the promotion of localization after the combination. Literature [23] emphasizes that the combination of Marxism and Chinese excellent traditional culture can inject new cultural impetus into the latter, thus resolving the internal contradictions of Chinese traditional culture in order to achieve cultural self-confidence and free development. It also stated that promoting the integration of Marxism and Chinese traditional excellent culture is an important way to promote and inherit Chinese traditional excellent culture. Literature [24] reveals that the combination of the basic principles of Marxism and Chinese traditional excellent culture is necessary to promote the development of the historical and cultural underpinnings of Marxism as well as to promote the transformation of Chinese traditional excellent culture into modernization.

The analysis of the integration path of Marxist education and Chinese excellent traditional culture is the basis for exploring the application of machine learning technology in Marxist education. Specifically, the CS algorithm is first improved using differential evolutionary algorithms to enhance its search ability and speed up its search speed. Then the improved CS algorithm is used to optimize the number of hidden layer nodes of the ELM structure, adaptively select the number of hidden layer nodes, and select the input weights and thresholds to complete the construction of the evaluation model of Marxist education based on ICS-ELM. The evaluation indexes of the effect of Marxist education integrating Chinese excellent traditional culture are selected, and the evaluation data of a university from 2016 to 2023 are used as samples to assess the educational effect, and the results of the prediction and actual value errors of different models are compared. On the other hand, based on the reinforcement learning mechanism in data mining to analyze user characteristics, combined with similarity calculation and k-means clustering algorithm to realize the resource model construction and resource recommendation, to establish the Marxist education resource recommendation model. The response time, MAE, and RMSE of different models are compared, and the model is applied to actual teaching to explore the effectiveness of the model in recommending Marxist educational resources.

2

Marxist Education Integrating Chinese Excellent Traditional Culture

Chinese excellent traditional culture is a historical treasure accumulated after more than five thousand years of development, and integrating it into Marxist education can not only guide students to deeply understand the profound connotation and ideological value of the excellent traditional culture, but also enrich the content of Marxist education, and promote the effective implementation of the fundamental task of establishing morality and educating people.

2.1

Mining the Essence

The vitality of China's excellent traditional culture is activated by the power of Marxist truth, and the identity of the two in terms of essence and essence requires in-depth excavation of the intrinsic essence of historical development and innovation with the concept of inheritance and development. Colleges and universities need to adhere to the combination of the basic principles of Marxism and the excellent traditional Chinese culture, integrate the ideological resources and cultural achievements into the teaching work, constantly awaken the cultural spiritual genes hidden in the hearts of college students, cultivate the cultural self-confidence of college students, further forge the sense of community of the Chinese nation, continuously enhance the cultural identity and cultural pride of college students, and further strengthen the confidence in realizing the great rejuvenation of the Chinese nation.

2.2

Recognizing the value of heritage

Colleges and universities should strengthen the excavation and elaboration of the core ideological concepts of Chinese excellent traditional culture in the process of carrying out Marxist education work, and guide college students to deeply understand the inheritance value of Chinese excellent traditional culture in the historical development and comprehensively comprehend the scientific and truthful nature of Marxist beliefs in the light of the current characteristics of college students' national cultural aspirations.

2.3

Guiding innovative development

On the issue of the high degree of fit of Chinese excellent traditional culture into the field of Marxist education, it is necessary for colleges and universities to combine the construction of the pattern of ideological and political education to promote the creative development and innovative transformation of Chinese excellent traditional culture in the work of Marxist education. On the one hand, colleges and universities should combine the concepts and ideas behind cultural self-confidence with Marxism, and explore the cultural fundamentals contained in Chinese excellent traditional culture according to its basic positions, views and methods. On the other hand, colleges and universities should actively explore the practical path of integrating Chinese excellent traditional culture into the Marxism curriculum, reinterpret and explain Chinese excellent traditional culture from the content to the form under the guidance of the concept of Marxism, and play a role in integrating it into the basis of the teaching of Marxism by means of creative transformation and innovative development, so as to truly realize the organic integration of the two.

3

Assessment of Marxist education based on ICS-ELM

The rapid advancement of machine learning technology opens up new possibilities for the innovative development of university education. After analyzing the practical progression of the integration of Chinese excellent traditional culture into Marxist education above, this chapter applies machine learning technology to the evaluation of Marxist education, establishes the ICS-ELM model, which is used to explore the teaching effect of the integration of Chinese excellent traditional culture in Marxist education, and sets up experiments to analyze the evaluation effect of the model.

3.1

Extreme Learning Machine Algorithm

Extreme Learning Machine (ELM) is a machine learning algorithm based on modern statistical theory, which better overcomes the defects of traditional neural networks. The Extreme Learning Machine model is fully connected between all layers, which solves the problem of neural networks having to adjust the required parameters multiple times, resulting in slow computation.

Now assume that there is N arbitrary sample (X_i,t_i),X_i = (x_i],x_i2,…,x_in),t_i = (t_i1,t_i2,…,t_im),n for input and output layer dimensions respectively. The hidden layer nodes are set to K. The specific single hidden layer feedforward neural network is denoted as: 1 $\sum_{j = 1}^{K} v_{j} g (X_{i}) = \sum_{j = 1}^{K} v_{j} g (w_{j} \cdot X_{i} + b_{j})$ v_j, w_j are the output and input weights respectively, b_j is the hidden layer bias, and g_j(x) is the activation function. For the purpose of minimizing the output error, equation (1) can be expressed as: 2 ${\begin{matrix} \sum_{i = 1}^{N} \sum_{j = 1}^{L} v_{j} g (w_{j} \cdot X_{i} + b_{j}) - t_{i} = 0 \\ i = 1, 2, \dots, N \end{matrix}$

Simplified representation: 3 $L = Λ H$ 4 ${[\begin{matrix} g (w_{1} \cdot X_{1} + b_{1}) & \dots & g (w_{L} \cdot X_{1} + b_{L}) \\ ⋮ & ⋱ & ⋮ \\ g (w_{1} \cdot X_{N} + b_{1}) & \dots & g (w_{L} \cdot X_{N} + b_{L}) \end{matrix}]}_{N \times L} \cdot {[\begin{matrix} v_{1}^{k} \\ v_{2}^{k} \\ ⋮ \\ v_{L}^{k} \end{matrix}]}_{L \times m} = {[\begin{matrix} T_{1}^{k} \\ T_{2}^{k} \\ ⋮ \\ T_{L}^{k} \end{matrix}]}_{N \times m}$ H is the output matrix of the hidden layer node, v is the output weight matrix, T is the desired output matrix, m is the number of output variables, and k is 1,2,3,…,m.

Since the input weights W_i and the hidden layer bias b_i of the ELM are set randomly, the output matrix of the hidden layer can be obtained after the 2 parameters are determined randomly, denoted by H , and then the output weights v are obtained by equation (5): 5 $v = H^{+} T$

3.2

The ICS-ELM model

To address the shortcomings of the basic ELM model, it is proposed to improve the ELM model by optimization algorithm.CS algorithm is an emerging heuristic algorithm, in order to enhance the search capability and speed up the search speed of the CS algorithm, it is proposed to improve the CS algorithm by differential evolution algorithm, and optimize the ELM algorithm by the improved CS algorithm (ICS), so as to enhance the accuracy and stability of the prediction.

3.2.1

CS Algorithm

Cuckoo algorithm is an algorithm based on the combination of cuckoo's living and reproduction habits and Levy's flight criterion. It has fewer set parameters, higher generalization, and better local and global optimization seeking ability. In order to simulate the cuckoo's survival mode, this algorithm proposes three assumptions: 1)

Each cuckoo's flight path is randomized, and only one nest can be randomly selected to lay one bird egg each time.

2)

The optimal objective function for calculating the health value of all bird eggs is used, and the one with the optimal health value is passed on to the next generation.

3)

The number of nests is fixed, and the host will have a fixed probability p (0~1) of discovering cuckoo eggs, and if discovered, the host will rebuild the nest or destroy the eggs.

Birds in nature are mostly characterized by Levi's flight, which is a random flight mode composed by low-frequency long-distance and high-frequency short-distance flight. Through the Levi flight can give the cuckoo's nest with the flight path on the update way for: 6 $x_{i}^{i + 1} = x_{i}^{'} + α \otimes L (ε) 1 \leq i \leq n$ $x_{i}^{'}$ is the location of the i rd nest at generation t, ε is the Lévy flight parameter, and α is the step factor.7 $α = 0.01 (x_{i}^{t} - x_{i}^{' best})$ $x_{i}^{' best}$ for the optimal nest location at generation t. The probability density function obeying Eq. (8) is applied to represent the Levy stochastic flight search path with respect to time t, i.e: 8 $L (ε) \sim η = t^{- β} 1 < β \leq 3$

According to equation (6) it can be seen that the cuckoo bird algorithm in the flight of the Levi's flight, the nature of the Levi's flight can make the search path transformed from one region to another, the randomness is very high, the search range and the diversity of the population is very large, leading to the algorithm in the operation of the convergence of the speed of the too slow, the search of the inefficiency of the problem.

3.2.2

Improved CS algorithm

In this paper, the CS algorithm is improved using the variation strategy of differential evolution algorithm to increase the population diversity of the CS algorithm, improve the global search ability and accelerate the convergence speed of the CS algorithm.

The DE/best/2/bin differential evolutionary algorithm is used with the following strategy: 9 $V = X_{b} + F (X_{r 1} + X_{r 2} - X_{r 3} - X_{r 4})$ where X_b is the best individual in the current population, four randomly selected individuals X_rk(k = 1,2,3,4) in the current population, and F is the scaling factor of DE. Equation (9) is introduced into the CS algorithm, and $X_{h}^{i}$ denotes the position of the kth nest at generation i.

Equation (6) becomes: 10 $l_{i} = X_{h}^{i} + \partial \oplus L (λ)$

Location Updates: 11 $l_{i} = {\begin{array}{l} X_{h}^{i}, f i n e s s (X_{h}^{i}) < f i n e s s (l_{i}) \\ invariability, f i n e s s (X_{h}^{i}) > f i n e s s (l_{i}) \end{array}$ CS Algorithm Lévy searches for updates after mutating bird nest locations.

Four bird nest locations $X_{t 1}^{i}$ $X_{2}^{i}$ $X_{t 3}^{i}$ $X_{t 4}^{i}$ are randomly selected in generation i, then: 12 $V_{i} = X_{b} + F (X_{r 1}^{i} + X_{r 2}^{i} - X_{r 3}^{i} - X_{r 4}^{i})$ where V_i denotes the mutated nest position, and F has a value in the range of [0, 1]. X_b denotes the optimal nest position in generation i, and $X_{n}^{i}$ denotes the random nest position in generation i. Calculate the fitness value of the mutated bird's nest and determine whether to update $X_{h}^{i + 1}$ or not: 13 $X_{h}^{i + 1} = {\begin{array}{l} l_{i}, f i n e s s (l_{i}) & < f i n e s s (V_{i}) \\ V_{i}, f i n e s s (V_{i}) & < f i n e s s (l_{i}) \end{array}$

Then it compares with the optimal bird's nest position of the previous generation and updates the current optimal position if it is better.

Compared with other optimization algorithms, CS algorithm possesses the advantages of simple structure and few control parameters. The improved cuckoo search algorithm (ICS) retains the advantages of the CS algorithm and also increases the diversity of the bird's nest location population, which enhances the global search capability and accelerates the convergence speed.ICS-ELM utilizes the bird's nest locations obtained from the ICS algorithm, rounds them up to the number of nodes in the hidden layer of the ELM, and obtains the input matrix and the hidden layer threshold.

3.2.3

ICS Optimization of ELM Prediction Models

Optimization of ELM using ICS algorithm, adaptive selection of the number of nodes in the hidden layer, and selection of the value of the losing human rights and thresholds, the proposed algorithm to optimize the ELM using ICS, the specific steps of the implementation of ICS-ELM are as follows:

Step 1: Data normalization process.

Step 2: Set the ICS algorithm parameters. Set the discovery probability parameter p_a , N initial positions $n e s t_{0} = [X_{1}^{0}, X_{2}^{0}, \dots, X_{N}^{0}]$ , the fitness value to $f b e s t_{0} = [Y_{1}^{0}, Y_{2}^{0}, \dots, Y_{N}^{0}]$ , and the maximum number of iterations max_it.

Step 3: Select the optimal bird's nest location $X_{h}^{i}$ of the previous generation, search the bird's nest location l_i according to Eq. (10), and determine whether to update it for l_i by Eq. (11). Obtain the variant bird's nest position V_i from equation (12). take the obtained bird's nest position as the number of hidden layer nodes of ELM, calculate the root mean square error (RMSE) of the sample data as its fitness value, which is determined by equation (13) $X_{h}^{i + 1}$ , and then update it by comparing with the last generation optimal bird's nest position.

Step 4: Generate a random number rand and p_a compare rand < p_a , then randomly search for bird's nest position and update to replace the worst position in the bird's nest, otherwise no change is made.

Step 5: If the termination condition is satisfied, stop the search, otherwise return to step 3.

Step 6: Select that bird's nest location with the smallest fitness value rounded to the nearest integer as the number of hidden layer nodes M of ELM , and output the corresponding input weights W, thresholds b, and output weights β. Establish the CS-ELM time series prediction model.

Compared with the CS algorithm and the ICS algorithm, the ICS algorithm is improved by the DE algorithm, and the CS algorithm 6 searches for updates and does not directly enter the next generation, but carries out a mutation process, thus increasing the diversity of the bird's nest populations and increasing the search range, which in turn enhances the global search capability. From the above steps, it can be seen that the ICS-ELM algorithm uses ICS to adaptively select the number of hidden layer nodes when the prediction accuracy is the best, and to select the value of the losing human rights and thresholds, thus improving the accuracy of each prediction and enhancing the stability of the model.

3.3

Empirical evidence and analysis of results

Based on the constructed assessment model of Marxist education integrating Chinese excellent traditional culture, evaluation indexes are selected to carry out the assessment of the effect of Marxist education in a certain school, and the accuracy of the model's teaching effect assessment is explored.

3.3.1

Evaluation indicators

In order to correctly and effectively evaluate the status of the effect of Marxist education in colleges and universities, following the principles of scientific, systematic, concise, objective, comprehensive, comparable and measurable selection of evaluation indexes, and in order to maximize the impact of each evaluation index, a system of indicators of the effect of Marxist education in colleges and universities is constructed. It mainly consists of three levels, namely the target level, the criterion level, and the element level. These levels contain 5 first-level evaluation indexes and 24 second-level evaluation indexes.

The 5 first-level evaluation indicators are the integration effect A, teaching attitude B, teaching method C, teaching ability D and teaching effect E of Chinese excellent traditional culture and Marxist education.

The integration effect includes four second-level indicators: ideological integration, practical integration, methodological integration, and value integration. Teaching attitude includes seven secondary indicators: preparing for class on time, starting and finishing class on time, classroom discipline, targeted questioning, encouraging questioning, listening to opinions, and answering patiently. Teaching methods include the three dimensions of heuristic teaching, multimedia use, and classroom practice. Teaching ability includes five secondary indicators: in-depth presentation of difficult points, summarization of key points, clear organization, handling of emergencies, and use of picture and animation examples. Teaching effect includes four dimensions of assignment quality, cognitive level, knowledge utilization and moral enhancement.

After constructing the evaluation indexes of Marxism teaching effect in colleges and universities, the scores and final scores of each evaluation index are obtained by expert scoring, after which the scores of each evaluation index are used as inputs of ICS-ELM and the final scores are used as outputs of ICS-ELM to establish the evaluation model of Marxism teaching effect in colleges and universities of ICS-ELM.

3.3.2

Data sources

The data of this study comes from the evaluation data of Marxist education effect of integrating Chinese traditional culture in colleges and universities from 2016 to 2023 in a university, and the maximum value method is used to standardize the data obtained. In accordance with the 1-9 scale method, the data on the evaluation index score of Marxism teaching effect in colleges and universities and the data on the evaluation score of teaching effect are obtained by comparing each evaluation index two by two. The score and final score of each evaluation index are shown in Table 1. The teaching effect of Marxism integration of traditional Chinese culture in colleges and universities is categorized into five grades: very good, good, average, poor and very poor, with corresponding scores of [4.2, 5), [3.4, 4.2), [2.6, 3.4), [1.8, 2.6), [0, 1.8). The teaching effectiveness of Marxist education integrating Chinese traditional culture in the sample colleges and universities ranged from 1.839 to 3.854, and was located in the evaluation level of “poor to good”, with the best teaching effectiveness in 2020 and 2022, both scoring above 3.4, which was the best for the teaching effectiveness of Marxist education. “Better” years. Among them, the integration effect A of excellent traditional Chinese culture and Marxist education in 2017 and 2020 is 4.796 and 4.728, which is a “very good” level of integration effect.

Table 1.

Each evaluation index score and final score

Index	2016	2017	2018	2019	2020	2021	2022	2023
A	2.265	4.796	3.964	1.182	4.728	1.167	2.588	1.206
B	4.282	2.276	2.659	1.86	4.196	2.065	4.131	2.453
C	1.819	2.685	2.373	3.825	3.656	1.574	2.348	2.047
D	3.215	1.679	1.028	4.162	2.537	0.868	4.260	1.488
E	1.432	0.949	1.495	4.532	2.615	3.632	4.198	3.105
Total	2.544	2.394	2.216	3.158	3.746	1.839	3.854	2.053

3.3.3

Analysis of results

Eight sets of data were obtained by expert scoring from 2016 to 2023, and the evaluation data from 2016 to 2019 were used as the training set and the evaluation data from 2020 to 2023 were used as the test set. The training set data are used to establish the ICS-ELM ideological and political teaching effect evaluation model of colleges and universities, and the test set data are used to test the correctness of the ICS-ELM ideological and political teaching effect evaluation model of colleges and universities. In order to highlight the advantages of ICS-ELM ideological and political teaching effectiveness evaluation model in colleges and universities, ICS-ELM is compared with PSO-ELM, GA-ELM and ELM. The evaluation of college Marxism's teaching effectiveness is measured by using the root mean square error (RMSE) and correlation coefficient (R).

The results of the evaluation of Marxism teaching effectiveness in colleges and universities are shown in Figure 1. The comparison of evaluation results of Marxism teaching effect in colleges and universities is shown in Table 2. According to the evaluation results of the teaching effect of ideology and politics in colleges and universities, it can be seen: 1)

From the overall point of view of the evaluation of Marxism teaching effect in colleges and universities, the evaluation scores of the ICS-ELM model for Marxism education in each year are closest to the actual scores, and the RMSE values of the training set and the test set are 0.007 and 0.036, which are lower than that of the comparison model, and the corresponding R-values are 0.964 and 0.984, which are higher than that of the other models, respectively. Therefore, using the ICS-ELM model, the evaluation of Marxism teaching effectiveness in colleges and universities is the best.

2)

The evaluation accuracy of ICS-ELM, PSO-ELM and GA-ELM is better than that of ELM, mainly because the algorithms of ICS, GA and PSO optimize the weights and bias involved in the ELM model, which makes the evaluation accuracy of the teaching effect of Marxism in colleges and universities using the ELM model greatly improved.

Table 2.

The comparison of the results of the evaluation results of Marxist teaching effect

Methods	Training set		Test set
Methods	RMSE	R	RMSE	R
ICS-ELM	0.007	0.964	0.036	0.984
PSO-ELM	0.018	0.955	0.040	0.965
GA-ELM	0.015	0.947	0.045	0.937
ELM	0.019	0.932	0.055	0.926

4

Recommendation of Marxist Educational Resources Based on Data Mining

Through the scientific and rational application of Marxism teaching resources, the recommendation method can collect and analyze teaching-related data, enhance the teaching effectiveness of Marxism courses to a certain extent, and enrich students' learning experiences. After evaluating the effect of Marxism education, this chapter proposes a Marxism education resource recommendation method integrating Chinese traditional culture based on the enhancement of the effect of Marxism education, combined with the machine learning algorithm of data mining, and experimentally analyzes the performance of the method.

4.1

Methods for recommending educational resources

4.1.1

Analyzing user characteristics

Using the reinforcement learning mechanism in the data mining algorithm, the collection of relevant data is carried out in accordance with different distribution steps, and the application background state of the collected Marxist teaching resources is set as s. Then, in t time, its application background state is st , which is able to reflect a certain data information in the application background state of the Marxist teaching resources. At the same time, set the student learning state as x, and the historical behavioral characteristics in t time as xt. Through the deep learning method, set the reward value as T , and the increase in the process as 1. Then the expression of the reward value is: 14 $T_{k + 1} = {\begin{array}{l} T_{k} \\ T_{k} + 1 \end{array}$ where T_k is the absence of behavior and T_k + 1 is the presence of xt behavior. In the iterative process, the historical behavioral characteristics of the user are extracted and the corresponding reward value parameters are calculated. The recommender system will keep the current reward value parameters unchanged to maintain stable recommendation results when the user lacks historical behavior data. However, when the user has historical behavior data, the recommender system will adjust the reward value parameters according to this data to better meet the user's needs. By setting the characteristic values of Marxist teaching resources in different states to the maximum reward value, the recommender system can provide the most appropriate resource recommendations according to the user's needs.

4.1.2

Resource modeling

Analyze the feature dataset of students based on their preference information. By clarifying the relationship between the core knowledge points of the course and the structure of the course design, labeling the attributes of the knowledge points in the core knowledge base, and representing the features by adding label attributes, the similarity degree of the labeled Marxist teaching resources is calculated, and the near-neighboring Marxist teaching resources of the new resources are obtained. The formula for its similarity is: 15 $s i m (X, Y) = \frac{| S x \cap S y |}{| S x \cup S y |}$ Where Sx, Sy is different Marxism teaching resources. Calculating the similarity can get the label similarity between the Marxism teaching resources, sorted according to the frequency of the occurrence of the label in the resources, and first select the label features of the resources similar to the nearest neighbors to get the correlation of different Marxism teaching resources to the label, calculated as Eq: 16 $s i m (x, n) = \frac{0.5 k - | c o u n t (n i) - 0.5 k |}{\log (| c o u n t (n i) | + 1)}$ where count(ni) is the number of labeled resources of Marxism teaching resources. The label with the highest frequency of occurrence is obtained after sorting, and it is recommended as the highest relevance label of Marxism teaching resources to the corresponding new resources, so as to enrich the characteristics of Marxism teaching resources.

4.1.3

Inter-domain resource recommendations

By applying the k-means clustering algorithm, resources with similar degrees of interest are grouped into different clusters. In each class of clusters, the resources with the highest degree of similarity are found to provide more accurate recommendation results. Setting the clustering center point involves all the students. The distance between the center point and the students is calculated using the formula: 17 $r = \arg \min \sum_{i}^{k} ‖ u - C ‖^{2}$ where r is the nearest student category and C is the centroid. The cluster with the closest distance is selected as the result and the clustering center point is updated. When the algorithm converges, according to the actual situation of students, predict the target students' ratings of Marxism teaching resources, which is calculated by Eq: 18 $h = R + \frac{\sum [r s i m {(x, n)}^{*} R]}{\sum r s i m (x, n)}$ where R is the average student interest in all rated category attributes and r is the student category. Predict the target student's ratings based on the student's historical data. Sort the predicted results, match the recommended learning resource features with the student's features, and adjust the dynamic feature weights at any time to fit the student's needs. After obtaining the final weight values, the minimum value of the objective function is obtained and the sequence of recommended resources is output, thus completing the recommendation.

4.2

Model performance experiments

After the design of the Marxist teaching resources recommendation model based on data mining is completed, it is necessary to test the function of the model to verify the comprehensive performance of the model and to find and adjust the possible defects. The specific model performance experiments are listed below.

4.2.1

Experimental preparation and program design

In the hardware configuration of 3.25 GHz CPU 64 GB RAM, 1 TB hard disk servers 3 were used as the application, database, and Spark computation servers respectively for the system implementation. Under the operating environment of Windows 10.2, the designed functional part of the recommendation model was programmed and implemented using Java language.

In the model testing, the Marxist educational resources recommendation model based on data mining is compared with the teaching resources recommendation system based on knowledge graph technology (RRS 1) and the teaching resources recommendation system based on kernel-typical correlation analysis (RRS 2) on the same indexes. The comparison index selects the difference between the recommended results and the preset ideal recommendations when each recommendation system recommends Marxist teaching resources to different users.

4.2.2

Test data and analysis

The concurrency testing tool WRK is used to test the stress response of each resource recommendation system in an offline state. The comparison results of the average response time of the systems under the same concurrency are shown in Figure 2. The response time of this paper's model under different concurrency is 125~210ms, which is significantly lower than the other two resource recommender systems, and this paper's Marxist education resource recommendation model based on data mining has higher recommendation efficiency.

Sixty percent of the manually generated dataset of learners' preferences for teaching Marxism was recorded as the training set, while the remaining 40% was recorded as the test set. After training the various parameters of the recommender system using the training set, the three teaching resource recommender systems performed the statistics of mean absolute error (MAE) and root mean square error (RMSE) of Marxist resource recommendation based on the test set, and the results of the mean absolute error and the root mean square error of resource recommendation are shown in Table 3. The average absolute error (0.558~0.677) and root mean square error (0.751~0.823) of the recommendation of the data mining-based Marxist education resource recommendation model are slightly smaller than other systems when the recommendation is based on relevant data. It shows that the model presented in this paper is more accurate in recommending Marxist education resources.

Table 3.

The MAE and RMSE results of resource recommendation results

Group	Our model		RRS 1		RRS 2
Group	MAE	RMSE	MAE	RMSE	MAE	RMSE
1	0.558	0.803	0.789	0.870	0.929	0.956
2	0.561	0.779	0.848	0.906	0.845	1.008
3	0.627	0.819	0.799	0.837	0.867	1.015
4	0.616	0.808	0.761	0.879	0.901	1.021
5	0.665	0.823	0.788	0.909	0.868	1.018
6	0.586	0.791	0.757	0.894	0.824	0.911
7	0.677	0.751	0.795	0.975	0.918	1.043
8	0.634	0.787	0.785	0.864	0.823	0.977
9	0.652	0.758	0.818	0.916	0.906	1.001
10	0.624	0.769	0.792	0.942	0.882	0.992

4.3

Application examples

In order to verify the effectiveness of the Marxist educational resources recommendation model designed in this paper, this model was applied to the teaching in a university. After 50 teachers and 80 students were applied for 30 days respectively, a survey was carried out by means of questionnaires. 130 questionnaires were sent out on 2 occasions, 122 questionnaires were retrieved, and 122 questionnaires were valid. The collected data were analyzed and statistically analyzed using SPSS.

The comparison of the effect evaluation of the teaching resources recommendation model is shown in Table 4. After applying the recommended model, the degree of understanding, cognition, model characteristics and interest of teachers and students increased significantly (P < 0.01), which can be seen that there are obvious changes in the evaluation results before and after applying the recommended model of Marxist educational resources, and teachers and students have their own understanding of the model, and use the model according to their own needs, and at the same time make the model have a personalized characteristics.

Table 4.

Comparison of the effect of teaching resource recommendation model

Item	Before applying	After applying	t value	P value
Degree of understanding	2.22±0.35	2.69±0.36	-9.554	0.003
Cognition of teaching resources	2.56±0.52	2.62±0.53	-3.615	0.001
Model characteristics	3.14±0.38	3.57±0.36	3.566	0.008
Attention concentration	3.24±0.52	3.65±0.53	1.628	0.058
Study interest	2.56±0.61	2.77±0.63	2.591	0.004

5

Conclusion

The development of intelligent technology has an important impact on Marxist education in colleges and universities. The study applies machine learning technology to Marxist education that integrates Chinese excellent traditional culture, proposes an assessment model of Marxist education based on ICS-ELM and a recommendation model of Marxist education resources based on data mining, and designs experiments for the comparison of relevant indexes of different models to verify the actual performance of the models. 1)

The effectiveness of Marxist education in the sample colleges and universities is in the range of “poor to good”, with scores ranging from 1.839 to 3.854. In two of the years, the integration of Marxist education and Chinese traditional culture was “very good”, with a mean score of 4.7 or more. In the training set and test set, the RMSE values (0.007 and 0.036) and R values (0.964 and 0.984) of this paper's teaching effectiveness assessment model achieve superior results, reflecting the more favorable Marxist education assessment effect of this paper's model.

2)

In terms of response speed and resource recommendation accuracy, the Marxist teaching resource recommendation model in this paper performs better, with a response time within 210ms, and the MAE (0.558~0.677) and RMSE (0.751~0.823) are also much lower than that of the comparison model, and in the example application, it promotes the enhancement of the degree of cognition and learning interest of teachers and students. The recommendation model for Marxist education has a better effect on recommending teaching resources.

3)

The application of machine learning technology in Marxist education in colleges and universities can not only improve the quality and effect of education, but also provide new ideas and methods for the upgrading and innovation of Marxist education in colleges and universities, which is conducive to the cultivation of high-quality, socially responsible and comprehensive talents.

Language:: English

Publication timeframe:: 1 times per year
Journal Subjects:: Life Sciences, Life Sciences, other, Mathematics, Applied Mathematics, General Mathematics, Physics, Physics, other

Journal RSS Feed

Research on the Application of Machine Learning for Marxist Education Integrating Chinese Excellent Traditional Culture

Handong Wang

Published Online: Feb 03, 2025

Received: Sep 13, 2024

Accepted: Dec 28, 2024

DOI: https://doi.org/10.2478/amns-2025-0006

KeywordsCuckoo search, Extreme learning machine, Data mining, Machine learning, Marxist education

© 2025 Handong Wang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Keywords
Cuckoo search, Extreme learning machine, Data mining, Machine learning, Marxist education