Analysis of the Application and Practical Effect of AI Technology in Civic Education Management in Colleges and Universities

At present, the network has become the largest environment for teachers and students in colleges and universities to study and live, and it is also the largest variable faced by education. Under the network intelligence environment, the rapid development of blockchain, Internet of Things, artificial intelligence, cloud computing and other cutting-edge technologies, some of the intelligent technologies can complete the humanization of teaching behaviors, the naturalization of human-computer communication, and the agentization of complex tasks, and the intelligent services have been widely applied to economic, cultural, social and other Intelligent services have been widely used in all aspects of economy, culture, society and so on, and have become a powerful basis for improving the effectiveness of work. Let science and technology know you better. Science and technology have never had such a profound impact on national development and people’s lives as today. Intelligent service means that under the technical support of intelligence, artificial intelligence, 5G and other technologies, through the analysis of user behavioral data, real-time judgment of the user’s mind (expectations/needs) behind the behavior, the first time to actively provide users with guidance in line with their expectations, to help users get a better sense of experience and a higher sense of satisfaction from the collection of information, identification of the scenario, the perception of the atmosphere, to the linking of the community, to improve the environment, college students’ ideological and political From collecting information, identifying situations, perceiving atmosphere to linking communities and improving environment, the benign ecological reconstruction of ideological and political education of college students can be effectively carried out by relying on intelligent technology and environment, which can further stimulate college students’ enthusiasm for learning and inner vitality, and really make ideological and political education focus on the educational goal and educational significance of fostering socialist builders and successors with the help of intelligent services.

Based on the actual needs of teaching and management work, it is urgent to create an intelligent education management system for colleges and universities, combining data analysis methods and decision support theory to build a model of “Intelligent College Teaching Management System”. The system can provide intelligent data analysis functions, decision support functions, teaching management in colleges and universities have many decision-making problems. Decision support system can effectively improve the scientific and fairness of decision-making. A decision support system is shown in Figure 1.

Figure 1.

Decision support system component

Based on the basic data generated in the process of teaching and learning in colleges and universities, a data warehouse is constructed to track and analyze a large amount of data, such as course offerings, students’ course selection, students’ evaluation, and students’ grades, without interruption. Intelligent college teaching management system data use

Faculty data: analyze the data of the school’s faculty, and analyze the reasonableness of the school’s teacher structure through difference comparison, trend analysis, and index analysis. Provide support for the rationality of the construction of the faculty, the introduction of talent, talent ‘training and so on.

Teaching effect: including the analysis of performance, four or six grades, computer grade, graduate school, graduation, etc., to provide support for the school to identify teaching indicators that need to be strengthened. Professional data.

Analysis of professional status, including analysis of the status of each major, configuration, team structure, discipline construction, to provide support for the school’s professional development and construction.

Curriculum data: analysis of the curriculum including course distribution, textbook situation, student popularity, etc., to provide support for the school’s curriculum arrangement.

Practical Teaching: Analysis of practical teaching, including scientific research, awards, and internships to provide support for the rationality of the school’s practice and internship arrangements and allocation of funds.

Teaching management: analyze the status of teaching management, including the structure of teaching management team, teaching and research activities, exams and examinations, teaching order and quality control. Provide support for the effectiveness of the school’s teaching management. Teaching hardware data. Analyze the teaching hardware conditions, through various indicators, analyze the degree of configuration and structural rationality of teaching facilities, and provide support for the reasonable construction of teaching facilities.

We can simply think that intelligent education management is the process of “data production, data collection, data storage, data mining, data analysis, and value output”, and the production of “valuable” information as the basis for decision-making is that our work is efficient and intelligent, which is also the significance of AI technology and the foundation of artificial intelligence in the future. 1)

Data production The school’s intelligent management (school registration, personnel, finance, logistics, assets, one-card, network marking, monitoring and resources), school-level Internet of Things, mobile public platforms, cell phones, tablets, PCs, network behavioral logs, and other platforms can be the base of data production, and the data production of these platforms is directly related to the daily operation of the school, and the production of these data accumulates day by day without knowing it.

Regarding the difference, in other words the degree of dissimilarity, there are several ways of calculating it as follows: 1)

For scalar quantities, i.e., numbers with no directional significance, the commonly used ways of calculating the degree of dissimilarity are the Euclidean distance and the Minkowski distance: 1d(Xi,Xj)=(xi1−xj1)p+(xi2−xj2)p+…+(xin−xjn)pp

The formula for the Minkowski distance, when p = 2 and p = 1, can be thought of as the Euclidean distance and the Manhattan distance, respectively. The Euclidean distance is the aggregate distance between two elements in Euclidean space, and is most widely used because of its simplicity and interpretability: 2d(Xi,Xj)=(xi1−xj1)2+(xi2−xj2)2+…+(xin−xjn)2

This metric is obtained by calculating the ratio of the between-class variance to the within-class variance, the larger the CH, the tighter the classes themselves and the more dispersed the classes are from each other, which implies better clustering. Its calculation formula is as follows: 8CH=tr(Bk)tr(Wk)*N−kK−1 9Bk=∑qnq(Cq−c)(Cq−c)T 10Wk=∑q=1k∑x∈Cq(Cq−c)(Cq−c)T k denotes the number of clusters, N denotes the number of samples, C_q is the center point of the q th cluster, n_q denotes the number of samples in the q th cluster, c denotes the center value of the samples, B_k denotes the intraclass variance, W_k denotes the interclass variance, and t_r( ) denotes the trace of the matrix, which is the Euclidean distance from all the data points of the qth cluster to the center point of the q th cluster in this formula. This clustering evaluation method is computationally faster and more efficient than the profile coefficients, and is therefore suitable for the evaluation of clustering methods under large samples.

The DB value is obtained by dividing the sum of the average intra-class distances of any two classes by the center distance of the two classes and finding the maximum value. As can be seen from the formula, the minimum value is 0. The smaller the value, the closer the sample points within the same cluster and the more separated the sample points of different clusters. Since this method uses the European distance calculation, it is not suitable for clustering of circularly distributed data.

The dataization of student management in institutions of higher education has achieved certain results, for example, we strongly support the construction of school informatization, which provides a solid foundation for the dataization of management in institutions of higher education. However, there are also many problems in the dataization of student management in institutions of higher education, such as insufficient information collection, low utilization rate and so on. The reasons for these problems are also multifaceted, and this paper believes that they mainly include the lack of intelligent talents and the imperfect construction of the regulatory system.

In the era of intelligence, the popularization of artificial intelligence ideas continues to affect all walks of life. The management of students in institutions of higher education has also appeared opportunities and challenges, in this case, whether the original management of institutions of higher education to adapt to the needs of the times and be able to adapt to the development of the times has become an important point of concern. The purpose of the investigation in this paper is to understand the status quo of student management dataization in institutions of higher education in the context of intelligence, so as to make targeted and practical suggestions on the problems existing in the dataization of student management in institutions of higher education.

In order to better improve the content of this paper, this paper on the X higher education students to carry out a relevant questionnaire survey, the questionnaire is issued in the form of network, for the students of the major colleges and universities. The questionnaire was issued 300 copies, retrieved 285 valid questionnaires, the questionnaire validity rate of 95%. The survey object as shown in Table 1, the number of men is 160, the number of women is 125, the ratio of men and women is more balanced.

The reason why schools do not collect enough information about students is to a large extent directly related to the schools’ own inadequacy in building an intelligent platform for student management. The lack of a reasonable platform for communication between students and academic staff has, to a large extent, caused obstacles to the collection of student management data. This paper shows in the survey of students that when you encounter problems during school, you will solve them or seek help through that form. The results of the survey are shown in Table 2, college students will choose parents (220), classmates or class officers (178) occupy the top two. Only 29 students chose academic workers, only 40 students chose counselors, 14 students chose teachers, and 34 students chose the campus network, which also reflects that the implementation of the school’s academic staff for student management is not in place. The reason why students do not seek help from academic workers, counselors and teachers is largely related to the lack of a good communication platform between students and academic workers, counselors and teachers.

The use of AI technology in student management involves less scope, which has become an important factor hindering the dataization of student management in higher education institutions. In the survey of this paper, “What do you think AI technology can be mainly embodied in higher education institutions”, it can be seen that the majority of students believe that the current intelligent technology is mainly used in personalized learning and counseling (online teaching) (281), followed by the establishment of a complete student record (232).

In this paper, we obtain first-hand information about students in X university through collection and survey methods, and the data are strictly encrypted and decrypted before they are organized to ensure that the data are safe and do not touch the privacy of individuals. Specifically, the obtained data includes about 18,000 undergraduate students in three grades (i.e., 2021, 2022, and 2023) of a university in Beijing. The data attributes mainly include initial data at the beginning of enrollment (e.g., gender, grade placement, etc.), students’ Civic and Academic Development, and awards, etc. More detailed data attributes are shown in Table 3. The educational data generated during the students’ undergraduate period is variable and traceable, and through data mining methods, it is possible to gain a deeper insight into the educational laws hidden behind the data. The dataset including 10,300 students and 19 attributes.

The results are shown in Fig. 2, when K=3, the cluster profile coefficient is 0.41223, which is the best among all the cluster numbers, indicating that K=3 is the optimal number of clusters for the object of study in this paper. Therefore, the number of clusters of the K-Prototype algorithm was set to 3 in the study. Finally, three groups of students were obtained, labeled as the first group (G1), the second group (G2), and the third group (G3), with 4,122, 2,482, and 3,696 students in each group, respectively. The group characteristics of the three groups of students obtained will be specifically analyzed below.

Figure 2.

Silhouette coefficient of each cluster when K is in [2, 10]

4.2.1

Analysis of basic information

The characterization of the three student groups in terms of basic information is shown in Figure 3, in which the basic information includes students’ gender, ethnicity, political identity, candidate category, filing score, and family economic status. First of all, the highest value of the ethnicity attribute (X2) in the figure shows that the G3 group has the largest number of ethnic minority students. This indicates that in this dataset, ethnic minority students are dispersed among the groups with excellent performance in Civics and Politics study, medium performance in Civics and Politics study, and poor performance in Civics and Politics study, but the number of students dispersed in the group with poor performance in Civics and Politics study is the highest. Second, the attribute (X5) of the submission score does not show a great difference, indicating that the submission score is not very correlated with the Civic and Political Learning Achievement during the university period, which is contrary to the traditional empirical perception. Thirdly, group G2 obtained the lowest value in attribute X7, which means that students in the group with medium performance in Civics and Politics study have the least number of students with financial difficulties. This implies that most of the students in the medium Civic and Political Learning Achievement group in this school come from non-financially disadvantaged families, i.e., more favorable family conditions did not lead to better Civic and Political Learning Achievement.

Figure 3.

The differences between the three groups of students

4.2.2

Characterization of honor

The characterization of the three student groups in terms of honors received during university is shown in Figure 4. The influence of honors received by students during university (X8 whether they received school honors, X9 whether they received provincial or higher honors, and X10 the total amount of scholarships received) on students’ academic performance is shown in the figure. As can be seen from the figure, there is a strong correlation between students’ Civics and Politics academic performance and honors received in this dataset, and students in the group with excellent Civics and Politics academic performance have significantly more advantages in all levels of honors. A certain number of students in the group with medium academic performance have also won school and provincial honors. Specifically, group G1 in the group of excellent academic performance in Civics and Politics has the largest share of the number of honors, and among them, the largest difference between the groups is the total amount of scholarships X10, which numerically shows a significant imbalance between the groups.

Figure 4.

The three groups of students received an honorary difference

4.2.3

GPA Characterization

The characterization of the GPA of the three groups of students in each semester during their university study is shown in Figure 5. It can be found that the students’ performance in Civics and Political Science during their university study has a strong trend of retention, and the academic level of the three groups of students in each semester of the university is stable, especially in the freshman year, which directly affects the academic level of the following three years. Very few students’ performance fluctuated greatly, a phenomenon that does not exclude the interference of external factors such as unexpected events. It should be noted that this conclusion is different from the above conclusion that “there is no significant correlation between the grades in the entrance examination and the performance in Civics and Political Science during the university years”. The results of this study show that students’ high school grades do not necessarily have a direct relationship with their academic performance in college, but there is a strong correlation between students’ Civics performance in the first year of college and their Civics performance in the next three years, i.e., the Civics performance in the first year of college has a direct impact on the Civics performance throughout the entire college period. The Civics and Political Science learning achievement of the three groups of students in their freshman year, and the situation of the group characteristics of the three groups of students X19 intuitively confirms the above research results.

Figure 5.

The average gpa difference between the three students

By analyzing the regular characteristics of the three groups of student groups under the optimal clustering in terms of their submission scores, academic levels in each semester during college, family economic status and other important influencing factors, it is important to explore the behavioral patterns and behavioral rules of students and to improve the scientificity of educational decision-making. The study found that students can be divided into three groups: high Civics learning achievement, medium Civics learning achievement, and low Civics learning achievement, and the analysis of the characteristics of the three groups of students found that there is little correlation between the students’ high and low academic achievement during the university period and the admissions scores of the college entrance examination. There is a strong trend of maintaining students’ academic achievement during college from freshman year. There is no inevitable relationship between high and low academic performance and family economic status.

The questionnaire survey was conducted with 70 students (35 in the experimental class and 35 in the control class) who participated in the experiment, and the questionnaire survey was conducted in the way of on-site distribution and on-site recovery, so as to ensure the recovery rate of the questionnaire. In this study, two groups of experiments were conducted to study the teaching effect of applying intelligent Civic Education Management, and one experimental group and one control group were set up. Classes 1 and 3 of the Civics discipline in X university were selected to carry out the empirical study.

In order to further test whether the experimental group’s achievement progress is significant from the pre-test to the post-test of the experiment, this study uses the SPSS tool to compare and analyze the data of the experimental group’s pre-test and post-test achievements using one-way analysis of variance (ANOVA), and the results of the test of variance of the experimental group’s pre and post-test Civics achievements are shown in Table 4. From the table below, it can be seen that the average score of the experimental group in the pre-test is 70.13, and the average score of the post-test is 92.25, and it can be seen that the average score of the experimental group has been significantly improved. Through the one-way ANOVA test, it was found that the P-value was less than 0.01. Statistically, when the P-value is less than 0.05, it means that there is a significant difference, and when the P-value is less than 0.01, it means that the difference is very significant, which indicates that in this experimental study, the experimental group’s grades were significantly improved from the beginning of the semester to the end of the semester, reflecting that the intelligent management of Civic Education can help to promote students’ Civic learning.

In the school-level unified examination at the end of this Civics semester, the experimental class scores are higher than the control class, and the significance level P-value is 0.004, which is less than 0.01, indicating that the average scores show a significant difference, i.e., the experimental students’ final examination scores are significantly higher than those of the control class. At the same time, the experimental class students’ final compared to the beginning of the school year results also showed significant improvement. By comparing the findings of the experimental and control classes, the obtained mean scores and P-values indicate that the experiment has achieved good teaching results, i.e., intelligent Civics education management can significantly improve students’ academic performance in Civics.