Research on Information Technology Integration and Teaching Process Optimization in Digital Transformation of Civic and Political Education in Colleges and Universities

With the rapid development of information technology, the field of applying information technology to realize digital transformation is also expanding. In the process of opening up information systems in various industries, there is no unified standard, which brings certain difficulties to the integration of information systems. In different information systems, data interaction, data conversion, and other aspects of certain problems can prevent compatibility between information systems, which can lead to a weak information system.Compatibility between information systems in various industries can improve work efficiency, provide better service for users, and enhance the service functions of various industries.

The trend of applying information technology to realize digital transformation has also inevitably entered the field of university civic education, and how to realize the integration of various information technologies and systems in the digital transformation of university civic education has also become a hotly debated topic in the industry.

In view of the integration methods of different information technologies and systems in the digital transformation of the Civic and Political Education in colleges and universities, this paper mainly proposes the methods of data integration and algorithm integration.

With the purpose of optimizing the teaching process in the digital transformation of Civics education in colleges and universities, the next step of this research will be to build up the Civics Teaching Management System based on MVC framework and the Civics Personalized Online Learning System based on Python, respectively. Both will achieve information technology integration and enhance system performance by adopting the data and algorithm integration method with a common student user dataset and the use of collaborative filtering optimization algorithm.

The design of the Civics Teaching Management System consists of two main aspects, which are the module structure design and function module design of the system. The module structure design is shown in Figure 1. The learning function includes material uploading, content guidance, material indexing and hotspot ranking, the management function includes material uploading and changing, column changing, material categorization and forum management, the interaction function includes online interaction, learning forum, e-mail and video retrieval, and the record function includes learning archives, personal information, process records, communication records and insights. Functional module design is mainly based on the module structure, the Civics Teaching Management System is divided into nine functional modules, which are Independent Learning Area, Social Cognition Area, Learning Navigator, Teacher-Student Interaction, Teacher Style, Achievement Display, Data Download, Fine Products Zone and Learning Forum.

Figure 1.

Main module functions

NET platform to realize data sharing and network communication, mainly through the enterprise application in Castle to build enterprise application framework, the process of data access framework to the application framework, followed by the Web layer MVC framework and AOP, the process effectively reduces the difficulty of the development and reduces the development time [22].ActiveRecord in Castle will be each data table with each model class one-to-one correspondence, and the records in the data table correspond to the model instance, based on the object model instead of the database operation mode, so that the data and business logic layer access to independence, enhance the transparency of the system development, speak and accelerate the development progress. This paper applies the B/S three-tier architecture model in order to expand the program and realize the high security and flexibility of the system. B/S three-tier architecture for Civic Teaching Management in Higher Education [23].

The database will be MySOL. MySQL is an open source small relational database management system developed by MySQL AB, Sweden. Since MySQL is open source software, the software is completely free to install and use, easier to install and use, and supports multiple platforms at the same time, it is one of the most popular relational database management systems. Due to the advantages of MySQL databases, such as fast speed and ease of installation and use, MySQL is generally chosen as the database of choice for website development.

Meanwhile, the collaborative filtering algorithm proposed in this research is added to the system architecture to improve the recommendation efficiency of teaching management of Civics courses in universities. In addition, the system simplifies the process of data handling through ActiveRecord, which increases the stability and efficiency of the system.

Collaborative filtering recommendation algorithm can analyze the feature similarity of different users and recommend the resources that may be interested for the users, and the algorithm has lower requirements on the amount of data, which can better explore the potential resources that the users are interested in, and it is suitable for the personalized recommendation of the resources of the Civics Teaching Management System in colleges and universities. Firstly, the user’s interest model is created based on the user’s basic information, such as age, gender, academic performance, learning preference, etc., and the user-item scoring matrix can be obtained from the interest model, as shown in equation (1): (1) R(m,n)=[ r11r12⋯r1nr21r22⋯r2n⋯⋯⋯⋯rm1rm2⋯rmn ] \[R(\begin{matrix} m,n) \\ \end{matrix}=\left[ \begin{matrix} {{r}_{11}} & {{r}_{12}} & \cdots & {{r}_{1n}} \\ {{r}_{21}} & {{r}_{22}} & \cdots & {{r}_{2n}} \\ \cdots & \cdots & \cdots & \cdots \\ {{r}_{m1}} & {{r}_{m2}} & \cdots & {{r}_{mn}} \\ \end{matrix} \right]\] Where, r represents the score of users’ ratings of the items, and r_mn represents the mrd user’s rating of the nth item. The similarity between each user can be calculated by equation (1), when there is a higher similarity between users, it means that the more similar the Civics learning preferences between users. In this paper, Pearson’s correlation similarity algorithm is used to calculate the similarity between users, as shown in equation (2): (2) sim ( i,j) = ( ∑k∈rij (Rj,k−R¯j) (Rj,k−R¯i) )/(∑k∈rij(Rj,k−R¯i)2 ⋅∑k∈rij(Rj,k−R¯j) ) \[sim\;(\;i,j)\;=\;\left( \;\sum\limits_{k\in {{r}_{ij}}}{\;}({{R}_{j,k}}-{{{\bar{R}}}_{j}})\;({{R}_{j,k}}-{{{\bar{R}}}_{i}})\; \right)/\left( \sqrt{\sum\limits_{k\in {{r}_{ij}}}{{{({{R}_{j,k}}-{{{\bar{R}}}_{i}})}^{2}}}}\;\cdot \sqrt{\sum\limits_{k\in {{r}_{ij}}}{({{R}_{j,k}}-{{{\bar{R}}}_{j}})}}\; \right)\]

In the formula, k represents the Civics teaching resource, R_j,k represents the rating of user j on k, and R¯i${{\bar{R}}_{i}}$ and R¯j${{\bar{R}}_{j}}$ are the average ratings of i and j on a Civics teaching resource. Meanwhile, the cosine similarity algorithm is used to calculate the similarity between users and items once more, as shown in equation (3): (3) { im( i,j) =cos( i,j) =( i¯⋅j¯) /( ∥i¯∥⋅∥j¯∥) \[\left\{ \begin{array}{*{35}{l}} im(\;i,j)\;=\cos (\;i,j)\;=(\;\bar{i}\cdot \bar{j})\;/(\;\parallel \bar{i}\parallel \cdot \parallel \bar{j}\parallel ) \\ \end{array} \right.\]

In the formula, i¯$\bar{i}$ and j¯$\bar{j}$ are the vectors of the ratings of i and j on the Civics teaching resources. However, when using the cosine similarity algorithm to calculate the similarity, the user’s rating of the project resources is highly subjective, and there is no uniform standard for the rating of the project resources, which leads to a large error in the results of the similarity calculation, and the modified cosine similarity algorithm is used to correct the problem of the variability of each dimensional scale as shown in Eq. (4): (4) sim(i,j) =( ∑s∈rij(Ri,s−R¯i) (Rj,s−R¯j) )/(∑s∈rij(Ri,s−R¯i)2⋅∑s∈rij(Rj,s−R¯j) ) \[sim(i,j)\;=\left( \;\sum\limits_{s\in {{r}_{ij}}}{({{R}_{i,s}}-{{{\bar{R}}}_{i}})}\;({{R}_{j,s}}-{{{\bar{R}}}_{j}})\; \right)/\left( \sqrt{\sum\limits_{s\in {{r}_{ij}}}{{{({{R}_{i,s}}-{{{\bar{R}}}_{i}})}^{2}}}}\cdot \sqrt{\sum\limits_{s\in {{r}_{ij}}}{({{R}_{j,s}}-{{{\bar{R}}}_{j}})}}\; \right)\]

Using Pearson correlation similarity, we can calculate the similarity of the program resources preferred by a certain user, while the modified cosine similarity algorithm can also calculate the similarity between users. The collaborative filtering algorithm is used to intelligently recommend Civics teaching resources for users, but the algorithm has data sparsity and cold-start defects, such that it cannot better provide users with resources of interest, reduces the accuracy of resource recommendation, and the scoring matrix presents a more sparse state [24]. In order to better recommend Civics teaching resources for users and assist teachers and students in teaching and learning, this paper invokes the K-means clustering algorithm to improve the collaborative filtering algorithm [25]. Firstly, the distance between the pre-set initial clustering center and a certain point is calculated using the Euclidean formula, as shown in equation (5): (5) E(x,y) = ∑i=1n(rui − rui)2 \[E(x,y)\;=\;\sqrt{\sum\limits_{i=1}^{n}{{{({{r}_{ui}}\;-\;{{r}_{ui}})}^{2}}}}\] Where, x,y denotes the rating r_ui of the user u,v on the project resource, r_ui denotes the rating u,v on the project resource i, and n denotes the number of the project resource. After obtaining the average value of all category data points, the data point closest to the average point can be used as the new clustering center, as shown in equation (6): (6) ci = (∑x = Cix) /m \[{{c}_{i}}\;=\;(\sum\limits_{x\;=\;{{C}_{i}}}{x})\;/m\] where C_i is the data of the ind category, c_i is the mean point of the ith category, and m_i is the number of data points of the ith category. In this paper, the problem is optimized by using bisecting K-means No. method, where all the data points are treated as a cluster, and the clusters are divided by the sum of squared intra-cluster errors SSE until the number of clusters reaches a predefined value.The SSE is calculated as shown in Equation (7): (7) SSE = ∑i=1k∑x∈Cidist (ci ,x)2 \[SSE\;=\;\sum\limits_{i=1}^{k}{\sum\limits_{x\in {{C}_{i}}}{d}ist\;{{({{c}_{i}}\;,x)}^{2}}}\] where dist() denotes the E(x,y) between the vectors. Fusion of K-means and dichotomous K-means, the collaborative filtering recommendation algorithm can be improved. After the construction of the user-item rating matrix, the data are clustered and analyzed by the K-means algorithm, and the contour coefficient S(i) is used to derive the K value of the optimal effect of clustering. The S(i) calculation is shown in equation (8): (8) S(i)=(b(i)−a(i)) /(max{ a(i) ,b(i) } ) \[S(i)=(b(i)-a(i))\;/(\max \{\;a(i)\;,b(i)\;\}\;)\] where b(i) is the average clustering of sample point i with all data points in the nearest cluster, and a(i) is the average distance between sample point i and all other sample data points in the same cluster. The larger the value of S(i), the more desirable the distance clustering effect. And then get the most ideal K value, will be used as the bifurcated K-means algorithm of the loss of value, talk about the secondary clustering, according to the clustering results to the user has not rated, but similar to the user has rated the project resources for predictive scoring. The calculation of predictive scoring r^ui${{\hat{r}}_{ui}}$ is shown in equation (9): (9) r^ui = r¯u+ ( ∑v∈Nmsim(u ,v)(ru− r¯v) )/(∑ε∈Nm|sim(u,v)|) \[{{\hat{r}}_{ui}}\;=\;{{\bar{r}}_{u}}+\;\left( \;\sum\limits_{v\in {{N}_{m}}}{sim(u\;,v)({{r}_{u}}-\;{{{\bar{r}}}_{v}})}\; \right)/\left( \sum\limits_{\varepsilon \in {{N}_{m}}}{|sim(u,v)|} \right)\] where r^ui${{\hat{r}}_{ui}}$ denotes the predicted rating of u for item resource i, r¯u\[{{\bar{r}}_{u}}\] denotes the average rating of user u for all item resources, and N_m is the top m users with the highest similarity to the user. The user similarity is calculated by combining Pearson similarity and Jaccard similarity as shown in equation (10): (10) sim(u,v)=simJacoard(u,v)⋅simPearson(u,v) \[sim(\begin{matrix} u,v) \\ \end{matrix}=si{{m}_{Jacoard}}(\begin{matrix} u,v) \\ \end{matrix}\cdot si{{m}_{Pearson}}(\begin{matrix} u,v) \\ \end{matrix}\]

The system as a whole is modularized and divided into three modules: teacher, student, and administrator. Teacher module is mainly for teachers to use, teachers can log in the system with an account and password to upload high-quality course chapter video and multimedia courseware and other learning materials, and then give students a certain course learning tasks, students as long as the completion of each course unit learning tasks after the unit of knowledge test, the teacher can also be on the student’s answers and homework for correction.

Student module is mainly for student users to use the subsystem, student users can log in to the system to watch online course videos and courseware, course learning can be carried out after the end of the online examination, in the course of learning process encountered difficult problems, you can ask the teacher questions at any time in the system or with other students to discuss exchanges, which can effectively improve the initiative of the students to learn independently.

Administrator module is mainly for the system administrator to log in the background system, the system administrator has the highest authority of the system, the system administrator can not only effectively manage the teachers and students of the system to log in the account password and other information, and can set up the system access privileges of the teacher user, but also can be customized for all the functions of the system.

Database design is the basis for the development of database application system, when designing the database, the relationship between each module should be fully considered, to create the right size, low redundancy and data reading efficiency of the data model. The database of this system includes student information table, teacher information table, chapter information table, video information table, course information table, comment information table, test question information table and so on.

System throughput (TPS) is the amount of system transactions per second. In this section, the system throughput test is carried out respectively on the comprehensive system of information on Civics and Political Education in colleges and universities constructed in this paper, and the test results are specifically shown in Fig. 2. A1 and A2 in the figure represent the Civics Teaching Management System and the initial Civics Teaching Management System in this paper respectively, while B1 and B2 represent the Civics Personalized Online Learning System and the initial Civics Online Learning System in this paper. From the figure, we can see that the system throughput of the IT-integrated Civic and Political Teaching Management System and the Civic and Political Personalized Online Learning System is always higher than that of the non-IT-integrated system, and when the number of requests reaches the maximum of 1,000, the system throughput reaches 12.3 and 11.39 respectively, which is higher than that of the non-IT-integrated system by 11.18 and 6.65, respectively. This demonstrates the excellent processing capability of the system server.

Figure 2.

TPS

The goodness of the system does not only depend on whether its functions can be realized, but also closely related to the performance of the system. This section will test the data synchronization performance of this paper’s comprehensive system of information on ideological and political education in colleges and universities. Select the file size of 100M, 500M, 1G, 10G, 50G files, respectively, respectively, three transmission test, take the average time consuming, test results are specific as shown in Table 1. Along with the A1, A2, B1, B2 pronouns used in the previous section. As can be seen from the table, regardless of whether the system is integrated with information technology, the file transfer time increases with the increase in file size. However, a horizontal comparison between the comprehensive information system of college civic education constructed in this paper and the system without information technology integration shows that the file transfer time of the civic education management system and the personalized online learning system of civic education in this paper is significantly faster than that of the non-information technology integrated system in the comparison. The transfer time of 100MB, 500MB, 1GB, 10GB and 50GB files in this paper is faster than that of the initial ideological and political teaching management system (4.06s), 8.29s, 11.6s, 33.15s and 394.86s, respectively, while the ideological and political personalized online learning system in this paper is higher than that of the initial ideological and political personalized online learning system by 1.32s, 4.84s, 8.26s, 84.16s and 478.53s, respectively. Obviously, after the integration of information technology, the synchronization performance of the system has significantly improved.