Research on Digital Resource Construction and Intelligent Recommendation Strategy for Civic and Political Education in Colleges and Universities

Digital intelligence technology covers the process of intellectualization in the fields of scientific thought, scientific concepts and technology’s modes of thinking, modes of emotion, habits of behaviour, standards of value and methods of work. In addition to allowing people to acquire knowledge, skills, and practical experience in science and technology, digital intelligence technology also has a profound impact on society, economy and the environment. With the progress of science and technology and society, it is of great significance to utilize digital intelligence technology to carry out ideological and political education.

The ideological and political theory course is a key course for implementing the fundamental task of establishing moral education, and it is necessary to consolidate the foundation of the digital transformation and development of the Civic and political course through the digital construction of teaching resources and to provide important support for the enhancement of the quality and efficiency of Civic and political course teaching [1-3]. In the digital era, teaching resources have a new physical carrier through digital technology to achieve the mining, integration and construction of various types of teaching resources such as text, pictures, video, audio, data, etc., and gradually evolve into a collection of multiple forms of teaching resources and to provide data interactive application of digital information systems and network learning sharing platform [4-7]. Only by accumulating and integrating digital teaching resources, unified management and open sharing can we provide powerful data support for the big data analysis of teaching activities in Civics and Politics courses, cloud computing processing of teaching data, and even educational intelligence so that Civics and Politics courses can produce more extensive, deep and effective changes under the empowerment of digital technology [8-11].

An intelligent recommendation system, as an efficient resource filtering tool, enables students to obtain personalized Civics elements from the Civics resource library, enhances the autonomy of students’ online Civics education for themselves, and improves the effectiveness of online Civics education [12-13]. The intelligent recommendation system uses machine learning algorithms to analyze students’ learning history and real-time feedback in order to identify students’ knowledge blind spots and points of interest and then recommend corresponding learning materials and activities [14-15]. In addition, by collecting and analyzing students’ learning data, teachers can access students’ overall learning trends and individual differences, thus providing data support for teaching decisions [16-18]. Through accurate, personalized recommendations and in-depth data analysis, Civics teachers can better meet students’ learning needs, improve teaching quality, and cultivate qualified socialist youth for the new era [19-20].

This paper connects digital resource construction and Civics education through the concept of personalised learning to create a Civics education pattern in colleges and universities under network resources. The Civics course resource network is constructed and represented using binary groups to provide data support for subsequent mining of Civics learners’ behaviors and interests. A keyword set is established based on the historical search keywords of Civic and Political learners, and the forgetting factor is used to calculate keyword weights and establish the learning interest set. Using collaborative filtering technology, we analyse the degree of similarity between learners’ behaviour and resource usage preferences, and complete the personalised recommendation of Civics course resources in colleges and universities based on collaborative filtering. On this basis, the improvement of the hybrid recommendation algorithm is achieved by combining the content of resources. Through simulation experiments, the degree of recommendation matching of the proposed algorithm in this paper and the degree of students’ favorite are tested. The model is implemented, and the scale measures the teaching effect of the Civics course under the digital resource recommendation.

2

Digital Resources Construction for Civic and Political Education in Colleges and Universities

2.1

Construction of Network Resources for Civic and Political Education in Colleges and Universities

2.1.1

Personalised Learning

There is no uniform definition of the concept of personalised learning in the academic world, and some researchers define personalised learning as a kind of learning in which students put themselves at the centre of learning, adopt autonomous, exploratory, reflective and innovative learning methods for their personality characteristics and development potential, and promote each learner to obtain learning that is suitable for their full, free and harmonious development. According to some data research studies, personalized learning mainly talks about student autonomy, comprehensive development of students, and teacher guidance. The author defines personalised learning as follows according to the definition that has a high degree of relevance to the topic of this study: personalised learning refers to learner-centred learning, using differentiated resources, open platforms, diversified interactions, and adopting self-directed, guided and diversified learning methods in response to the student’s personality characteristics and development potential, and prompting each learner to obtain a kind of learning that suits their full, free and harmonious development. Personalised learning refers to a kind of learning that uses differentiated resources, open platforms and diversified interactions and adopts autonomous, guided and diversified learning methods for students’ characteristics and development potentials so as to encourage each learner to obtain their comprehensive development.

Personalised learning uses the advantage of resources to form a whole chain of educational ecology, which helps to create a large educational pattern, using the network to collect learning data so as to better meet the learning needs of students to achieve interactive communication, complete personalised learning, and achieve all-round development [21]. Personalised learning can cultivate college students’ creativity and guide college students to perfect their personality and transcend themselves, thus promoting college students to achieve full, free and harmonious development. On this basis, the concept of personalised learning is a concept based on the enrichment of human personality, i.e., the learner perfects the unique personality through learning, and the individual carries out learning according to their interests, qualities and characteristics. Psychological research shows that personality traits, inner strength, and other non-intellectual factors are the key to influencing people and leading them towards perfection. Due to the differences in the learner’s emotions, will, motivation, hobbies, and character, learning has a different impact on them. Personalised learning is built in the general environment of teacher-student interaction in education, the construction of a relatively stable structure of network resources and related operational procedures, students’ differences and different needs for learning are respected, students learn with good learning support to achieve all-round development.

2.1.2

Network Resources for Civic and Political Education in Colleges and Universities

Civic and political education resources in colleges and universities refer to all the civic and political education resources that are conducive to the promotion of civic and political education in colleges and universities and the achievement of the goals and tasks of civic and political education. According to some data surveys and analysis, college civic education resources are mainly divided into the following different categories based on different perspectives.

Civic and political education network resources in colleges and universities have a wider meaning. They refer to all the information resources that can be used for civic and political education in colleges and universities transmitted through the network. This covers all the network resources that can enter the classroom and the learning field of college students, and that can serve the activities of college civic and political education and help improve the quality of civic and political education. This study defines it as follows: all the information resources that can be used for the teaching and management of ideological and political education in colleges and universities in order to promote the perfection of the personality and comprehensive development of college students.

2.2

Content of network resources construction for higher education civic education

2.2.1

Main contents of the construction of the Civic Policy resources

The construction of network resources for ideological and political education in colleges and universities refers to the educational and teaching management way of ideological and political educators in colleges and universities to design, select, produce, apply, transmit, exchange and feedback the ideological and political education information according to certain norms based on the cognitive characteristics of students with the support of network technology in order to make the network information have a positive effect on the students and to make them cultivate their ideological and moral qualities in line with the needs of the society.

The main content of the network resource construction of ideological and political education in colleges and universities is based on the classification of resources, specifically in terms of the construction of curriculum resources, the construction of resource platforms, the use of network resources, and the construction of network learning interactions, including the following main contents.

First, the opening of the Civic and Political Network Course. Including ideological and moral cultivation, legal foundations, an outline of modern Chinese history, and other course subjects.

Second, utilising the form of network material. Including the use of the network above for learning resources, media materials, question banks, questions and answers, test paper materials, and resource catalog citation.

Thirdly, building a platform for network learning. Including the mentioned network learning platform, Civic Education network course learning platform, Civic Education theme website, and so on.

Fourth, utilise the network platform, including application software such as Learning Power, learning platforms such as Rain Classroom, Blue Ink Cloud Class, and so on.

Fifth, carry out the interaction of online learning in civic and political education. Including online and offline learning integration, as well as interaction within and outside the classroom network for learning.

Sixth, carry out online learning assessment for Civic and Political Education. Including closed-book theory exams, open-book theory exams, essay writing, personalized assessment methods, and so on.

2.2.2

Basic Principles of Network Resources for Civic and Political Education in Colleges and Universities

Firstly, taking students as the main body, we will broaden diversified learning channels. Educational activities cannot be separated from the two main bodies of teachers and students, and their purpose is to promote the growth and development of students. Traditional educational activities often focus on the teacher’s classroom organization and teaching, ignoring the development of individualized competence of the student as the main focus. If students have the right to choose, they should creatively design learning resources, learning tools and feedback, fully invoking the combination of the elements to provide college students with the maximum degree of autonomy diversified learning conditions.

Secondly, learning resources should be enhanced to encourage students’ independent construction. Some students will have unhealthy addictive psychology in the process of using network resources for learning, so teachers must screen out valuable learning resources and provide timely guidance to enhance students’ ability to search for information and construct on their own.

Thirdly, give full play to the role of civic and political education network resources and respect the individual development of students. Civic and political education network resource construction should be able to timely solve students’ ideological confusion, form spiritual leadership, and promote students’ individual development and personality perfection in a learning atmosphere of respect and equality.

3

Personalised Recommendation of Civic and Political Curriculum Resources in Colleges and Universities

3.1

Construction of Civics Programme Resource Network

In order to facilitate the subsequent mining of learning behaviours and interests, a binary group is used to represent the resource network of the Civics course, which is formulated as [22]: (1) $K = (H_{i j}^{n}, L_{g f}^{m})$ \[K=(H_{ij}^{n},L_{gf}^{m})\]

The directed graph of the trajectory of Civics course resource usage is: (2) $H_{i j}^{n} = (J, P)$ \[H_{ij}^{n}=(J,P)\]

The network of the Civics and Political Science course resources contains learning cognitive ability, knowledge structure, and learning preference, which is expressed as a triad: (3) $L_{g f}^{m} = (U, O, E)$ \[L_{gf}^{m}=(U,O,E)\]

In this paper, two vectors, $H_{i j}^{n}$ \[H_{ij}^{n}\] and $L_{g f}^{m}$ \[L_{gf}^{m}\], are used to construct the Civics Course Resource Network K, so as to present the trajectory of the use of Civics course resources, reflecting the type, frequency, and content of the use of Civics course resources, and to provide a basis for the mining of learning interests.

3.2

Learner Interest Mining

In the Civics Curriculum Resource Network, the keywords in the nst Civics Curriculum Resource Usage Trajectory are V. This paper establishes keyword sets based on learners’ historical search keywords and uses weights to update learners’ search keywords. In the process of mining interests, the formula for introducing the forgetting factor to calculate the keyword weights is: (4) $ϖ = K^{\frac{\ln (T - t)}{r}} + Y_{ν} \times ε \times γ γ$ \[\varpi ={{K}^{\frac{\ln (T-t)}{r}}}+{{Y}_{\nu }}\times \varepsilon \times \gamma \gamma \]

The weights of the top d learning interests are normalised according to the actual demand, and the learning interest set of the Civics course resources is constructed. The formula of this collection is: (5) $S = {ϖ_{1} s, ϖ_{2} s, \dots, ϖ_{d} s}$ \[S=\{{{\varpi }_{1}}s,{{\varpi }_{2}}s,\cdots ,{{\varpi }_{d}}s\}\]

Where: S is the interest in learning the resources of the Civics programme and s is the learning interest node.

3.3

Personalised recommendation based on collaborative filtering

The collaborative filtering technique is used for similarity analysis, assuming that the learner is x and the two resource use sequences of learner x in the resource use trajectory of a certain Civics and Politics course are e₁,e₂ respectively [23].

(6)

C_{L_{g f}^{m}} = S \times G Q + (1 - w) G Q

\[{{C}_{L_{gf}^{m}}}=S\times GQ+(1-w)GQ\]

Calculate the similarity degree between learners’ interest in learning Civics course resources and Civics course resources, the specific formula is: (7) $μ = \frac{2 C_{L_{s j}^{m}} e}{e_{1} + e_{2}}$ \[\mu =\frac{2{{C}_{L_{sj}^{m}}}e}{{{e}_{1}}+{{e}_{2}}}\]

According to Eq. (7), the Civics course resources are sorted in the order of big to small, the Civics course resources with similarity degree ranked in the first z are filtered according to the demand, and the Civics course resources that do not meet the demand are filtered out to establish the alternative samples of personalised recommendation, and the expression is: (8) $ρ (μ) = ξ = δ \times (μ - σ_{1} σ_{2})$ \[\rho (\mu )=\xi =\delta \times (\mu -{{\sigma }_{1}}{{\sigma }_{2}})\]

To ensure the accuracy of the recommendations, each resource in the recommended alternative set ρ(μ) is scored. The scoring is based on the importance of Civics course resources and learning needs as indicators, and the specific formula is: (9) $v = y_{ρ (μ)} \times φ$ \[v={{y}_{\rho (\mu )}}\times \varphi \]

Where: v is the comprehensive score of the resources in the alternative set of personalised recommendation of civic politics course resources ρ(μ), y_ρ(μ) is the usage rate of the ith course resource in the alternative set of personalised recommendation of civic politics course resources ρ(μ) in the network of civic politics course resources K, and φ is the value of the degree of usage of civic politics course resources in the network of civic politics course resources K.

3.4

Hybrid recommendation algorithm based on collaborative filtering and resource content

3.4.1

Selection of hybrid recommendation algorithms

Figure 1 shows the flow of the hybrid recommendation algorithm. In order to improve the real-time, accuracy and applicability of the recommendation model of the curriculum Civics, this paper proposes a hybrid recommendation algorithm based on collaborative filtering and resource content because the users of the curriculum Civics learning system are relatively fixed and have the same hierarchy. The number of resources with common ratings is large, so Pearson similarity is used as the method for similarity computation in the recommendation model. However, the method cannot solve the cold start problem caused by the sparse data of users and resources, so the collaborative filtering mechanism that introduces a penalty factor improves the Pearson correlation coefficient to make up for the shortcomings in data sparsity [24].

3.4.2

Similarity Calculation Improvement

Similarity improvement of collaborative filtering algorithm

This paper introduces two penalty factors on the original Pearson similarity calculation [25]:

The first step is to construct the user-resource scoring matrix.

In the second step, the Pearson correlation coefficient is used to calculate the similarity between the Civic resources. (10) $s i m (m, n) = \frac{Σ_{u \in U_{m n}} (r_{u m} - \bar{r_{m}}) (r_{u n} - \bar{r_{n}})}{\sqrt{Σ_{u \in U_{m n}} {(r_{u m} - \bar{r_{m}})}^{2} \times \sqrt{Σ_{u \in U_{m n}} {(r_{u n} - \bar{r_{n}})}^{2}}}}$ \[sim(m,n)=\frac{{{\Sigma }_{u\in {{U}_{mn}}}}({{r}_{um}}-\overline{{{r}_{m}}})({{r}_{un}}-\overline{{{r}_{n}}})}{\sqrt{{{\Sigma }_{u\in {{U}_{mn}}}}{{({{r}_{um}}-\overline{{{r}_{m}}})}^{2}}\times \sqrt{{{\Sigma }_{u\in {{U}_{mn}}}}{{({{r}_{un}}-\overline{{{r}_{n}}})}^{2}}}}}\]

The third step is to add the weights of the popular resources penalty factor, the course Civics learning system in this paper aims to personalise the recommendation of high-quality Civics resources that students are interested in, so it is necessary to reduce the impact of popular resources on the similarity, and appropriately increase the weight of the ratings of non-popular resources to achieve the accurate recommendation. (11) $θ = \sqrt{\frac{Σ_{i ϵ l_{u p}} r_{u i} \times r_{v i}}{Σ_{i ϵ l_{u}} r_{u} \times Σ_{i \in l_{p}} r_{v}}}$ \[\theta =\sqrt{\frac{{{\Sigma }_{i\epsilon {{l}_{up}}}}{{r}_{ui}}\times {{r}_{vi}}}{{{\Sigma }_{i\epsilon {{l}_{u}}}}{{r}_{u}}\times {{\Sigma }_{i\in {{l}_{p}}}}{{r}_{v}}}}\]

The fourth step is adding a time decay penalty factor, mainly for the user to browse the resources with the change of time and change the situation of the database course of the course of Civics, for example, the teaching progress with the advancement of time, the content of the students to learn and the technical points are also in-depth, and each chapter of the course of the Civics elements and the specific knowledge is closely related to For example, at the beginning of the semester to learn the overview of databases and the basis of relational databases, and other knowledge, the mid-term study of SQL language and database table operations, and the end of the semester is a hands-on session on database design. There is a big difference between the Civics resources that students browse and grade in these phases, so a penalty factor should be added to those resources that have too long a time between grades. The time decay weights are shown in equation (12). (12) $f (t) = \frac{e^{t_{u m} - t_{u n}}}{1 + e^{t_{u m} - t_{u n}}}$ \[f(t)=\frac{{{e}^{{{t}_{um}}-{{t}_{un}}}}}{1+{{e}^{{{t}_{um}}-{{t}_{un}}}}}\]

The similarity calculation formula after combining the two penalty factors is shown in equation (13): (13) $s i m_{1} (m, n) = θ \times f (t) \times \frac{Σ_{u \in U_{m n}} (r_{u m} - \bar{r_{m}}) (r_{u n} - \bar{r_{n}})}{\sqrt{Σ_{u \in U_{m n}} {(r_{u m} - \bar{r_{m}})}^{2}} \times \sqrt{Σ_{u \in U_{m n}} {(r_{u n} - \bar{r_{n}})}^{2}}}$ \[si{{m}_{1}}(m,n)=\theta \times f(t)\times \frac{{{\Sigma }_{u\in {{U}_{mn}}}}({{r}_{um}}-\overline{{{r}_{m}}})({{r}_{un}}-\overline{{{r}_{n}}})}{\sqrt{{{\Sigma }_{u\in {{U}_{mn}}}}{{({{r}_{um}}-\overline{{{r}_{m}}})}^{2}}}\times \sqrt{{{\Sigma }_{u\in {{U}_{mn}}}}{{({{r}_{un}}-\overline{{{r}_{n}}})}^{2}}}}\]

Similarity improvement based on content recommendation

Firstly, the weight values of labels on resources are defined, and the resource-label matrix is constructed as shown in equation (14). (14) $R = [\begin{matrix} w_{11} & w_{12} & \dots & w_{1 b} \\ w_{21} & w_{22} & \dots & w_{2 b} \\ ⋮ & ⋮ & ⋱ & ⋮ \\ w_{a 1} & w_{a 2} & \dots & w_{a b} \end{matrix}]$ \[R=\left[ \begin{matrix} {{w}_{11}} & {{w}_{12}} & \cdots & {{w}_{1b}} \\ {{w}_{21}} & {{w}_{22}} & \cdots & {{w}_{2b}} \\ \vdots & \vdots & \ddots & \vdots \\ {{w}_{a1}} & {{w}_{a2}} & \cdots & {{w}_{ab}} \\ \end{matrix} \right]\]

After constructing the resource-label matrix, the weight value of the label to the user is calculated as shown in equation (15). (15) $w = \frac{r (t_{m})}{r (u_{i})} \times \frac{b}{a}$ \[w=\frac{r({{t}_{m}})}{r({{u}_{i}})}\times \frac{b}{a}\]

Comparing the content attributes of the resources calculates the similarity between the resources as shown in equation (16). (16) $s i m_{2} (m, n) = \frac{U_{m} \times V_{n}}{\sqrt{| U_{m} | \times | V_{n} |}}$ \[si{{m}_{2}}(m,n)=\frac{{{U}_{m}}\times {{V}_{n}}}{\sqrt{|{{U}_{m}}|\times |{{V}_{n}}|}}\]

3.4.3

Score Prediction for Hybrid Recommendation Algorithms

Combining the improved Pearson similarity sim₁(m,n) using the penalty factor and the similarity sim₂(m,n) calculated based on the content attributes to get the resource similarity of the hybrid recommendation algorithm, as shown in equation (17).

(17)

s i m (m, n) = α \times s i m_{1} (m, n) + (1 - α) \times s i m_{2} (m, n)

\[sim(m,n)=\alpha \times si{{m}_{1}}(m,n)+(1-\alpha )\times si{{m}_{2}}(m,n)\]

According to the improved hybrid algorithm similarity calculation to find the set of nearest neighbours of the target resource, the predicted score of user u for resource m is shown in equation (18).

(18)

p_{u m} = \bar{r_{m}} + \frac{\sum_{n \in S (m, K)} s i m (m, n) \times (r_{u n} - \bar{r_{n}})}{\sum_{n \in S (m, K)} | s i m (m, n) |}

\[{{p}_{um}}=\overline{{{r}_{m}}}+\frac{\sum\limits_{n\in S(m,K)}{s}im(m,n)\times ({{r}_{un}}-\overline{{{r}_{n}}})}{\sum\limits_{n\in S(m,K)}{|sim(m,n)|}}\]

3.4.4

Recommendation algorithm evaluation metrics

Evaluation metrics can be used to evaluate the performance of recommendation algorithms, and this section uses recommendation precision to evaluate the recommendation function of algorithms.

The main indicators of recommendation precision are accuracy and recall, and the accuracy and recall are shown in Equation (19) and Equation (20), respectively. Where R_u and T_u represent the actual recommendation results and the predicted recommendation results, respectively: (19) $\Pr e c i s i o n = \frac{R_{u} \cap T_{u}}{R_{u}}$ \[\Pr ecision=\frac{{{R}_{u}}\cap {{T}_{u}}}{{{R}_{u}}}\] (20) $Re c a l l = \frac{R_{u} \cap T_{u}}{T_{u}}$ \[\operatorname{Re}call=\frac{{{R}_{u}}\cap {{T}_{u}}}{{{T}_{u}}}\]

3.4.5

Recommendation Effect Analysis

In order to test the rationalisation of this paper’s recommendation method, three groups are set up. The experimental group of this paper’s method is composed of two groups: control A and control B, which are respectively big data recommendation and collaborative filtering algorithms. The application of recommendation ability on real data is carried out, and the variable parameter is set to the number of Civics teaching resources recommended to students. The recall and precision of different recommendation methods were calculated using different parameters. In the parameter value of 0 to 10, the precision degree and recall rate of different groups are calculated, and the specific values are shown in Fig. 2, Fig. (a) is the comparison of the results of the precision degree, and Fig. (b) is the comparison of the recall rate.

As can be seen from the experimental results, with the change of parameters, the precision degree of the experimental group stays more stable, with a precision degree of more than 89%, and has the slowest decay rate among the three group results. Meanwhile, the recall of the experimental group is larger, and the improvement is fast, as seen in Fig. (b). When the parameter K=10, the recall reaches 89.96%, which is the highest recall of the three groups. After many experiments, it is proved that using the method of this paper for the recommendation of digital resources for Civic Education can keep the results at a considerable level, and all the gains obtained are better than the traditional methods. In the actual recommendation scenario, it can achieve better recommendation results.

In summary, the use of the resource recommendation method in this paper can improve the timeliness of the recommended content and make accurate recommendations for teaching resources in Civics and Political Science. In the process of user search, user preference characteristics are extracted, and the matched content is presented to the user, which enables students to browse online Civics and Political Science teaching resources and recommends the online Civics and Political Science teaching resources that students are interested in. The cold-start problem is solved by collecting users’ historical behavioral data through their self-selected labels and inputting new feature data into the model. Using the recommendation method in this paper can help students make better use of online resources for teaching Civics and Political Science, which can improve their learning efficiency.

4

Experimental test analysis

4.1

Data preparation

In order to ensure the accuracy of the data test, students within a provincial university were randomly selected, and 1,000 students were each selected as test subjects in the order of freshmen to seniors. Civic and political education resources were retrieved from the national resource library, and thousands of groups of data were randomly selected as recommendation samples in accordance with the lecture courses of different grades, using the method proposed in this paper to recommend intelligent civic and political course resources to students. Figure 3 shows the data samples of Civic and political education resources, sorted by the number of recommendations. The top three recommended resources are A9, A10 and A3, and the total number of recommendations is 1669, 1240 and 1090, respectively.

4.2

Matching Comparison Results of Different Recommendation Methods

Here, different recommendation methods are used to recommend the Civic Education resources so that they can be taught in different stages of university lectures. The selected resource samples are uploaded to the MATLAB test platform and connected to three groups of recommendation methods, respectively. Figure 4 shows the matching test results of each group of methods, and the columns in the figure are, from left to right, A1, A2, A8, A3, A4, A9, A5, A6, A7, and A10. As can be seen in Figure 4, the method of this paper is affected by clustering, and it can effectively integrate different types of resource data to achieve the personalised recommendation of Civic and Political Education resources and the number of recommendations for each grade ranks the first among the three algorithms. Taking the fourth year as an example, the recommended quantity of control A group is 1295.81448, 1278.84615, control B group is 1277.71493, 1277.71493, and the experimental group is 1310.52036, 1334.27602, respectively. The recommended results have the fitness and certain application value with Civic and Political Education courses of all grades.

4.3

Resource Favouritism Test under Intelligent Recommendation

In order to further analyze the effect of different recommendation methods, the selected college students are used as test objects. The duration of individual courses of the Civics education resources under each type is set to 45 minutes, respectively, to determine the degree of liking for the recommended resources by students of different grades. Fig. 5 shows the results of the statistics of the listening duration of different Civics education resources. This paper’s method for different types of Civics education resources recommended can be maintained in the 42-minute or more listening time, indicating that all grades of college students have a high degree of liking for the recommended resources and can accept the type of Civics education. The other two methods result in a listening time of approximately 20 minutes. It can be seen that the method of this paper can not only meet the good matching of the Civics education resources but also ensure the students’ enjoyment of the Civics course, with a better recommendation effect, which can be put into the actual Civics teaching for application.

5

Measurement and results of the evaluation of the effectiveness of teaching Civics recommended by digital resources

5.1

Preliminary analysis of evaluation scales

Table 1 shows the item analysis and factor analysis of the evaluation scale of the teaching effectiveness of college civics under the recommendation of digital resources, and the exploratory factor analysis of the scale was conducted respectively, and it was found that the indicators of the evaluation dimension of the teachers’ digital teaching effectiveness could be clustered to a higher degree into three factors, namely, the teaching satisfaction, the teaching sense of acquisition, and the teacher-student interaction perception, and the cumulative contribution rate of the factor variance was 94.4875%. The indicators of students’ digital learning effect evaluation dimension can be clustered into 3 factors: students’ identification, students’ access, and teachers’ and students’ interaction perception to a high degree, and the cumulative contribution rate of factor variance is 92.4985%. The indicators of big data teaching effect evaluation dimension can be clustered into 2 factors of student achievement evaluation and teacher teaching evaluation to a high degree, and the cumulative contribution rate of factor variance is 95.7856%.

Table 1.

Project analysis and factor analysis

Primary indicator	Evaluation of teachers’ digital thinking
Secondary indicator	Satisfaction	Acquired sense	Student interaction perception
Project analysis t test	0.000***	0.000***	0.000***
KMO		0.9482
Contribution rate /%		50.2562
Cumulative contribution /%		94.4875
Primary indicator	Student digital learning effect evaluation
Secondary indicator	Satisfaction	Acquired sense	Student interaction perception
Project analysis t test	0.000***	0.000***	0.000***
KMO		0.9157
Contribution rate /%		47.8565
Cumulative contribution /%		92.4985
Primary indicator	The combination of recommended teaching results
Secondary indicator	Student evaluation	Teacher evaluation
Project analysis t test	0.000***	0.000***
KMO		0.9245
Contribution rate /%		52.7486
Cumulative contribution /%		95.7856

To further verify the convergent validity of the scale, Estimate, AVE and CR were measured, respectively. Table 2 presents the combined reliability and convergent validity of the scale used to evaluate the effectiveness of teaching Civics in colleges and universities using digital resources. Further, through the validation, it can be seen that the factor loadings of each latent variable of the dimension corresponding to each topic are above 0.84, and some of them are above 0.9, which indicates that each of its latent variables corresponding to the topic to which it belongs is highly representative. In addition, the average variance extracted for each latent variable is greater than 0.8, and the combined reliability coefficient (CR) is greater than 0.9, which indicates that the convergent validity is more satisfactory. The results show that it is established and feasible to construct the evaluation of the teaching effect of college Civics under the recommendation of digital resources into the first-level indexes of three dimensions: the evaluation of teachers’ digital teaching effect, the evaluation of students’ digital learning effect, and the evaluation of the teaching effect of big data.

Table 2.

Analysis of composite reliability and convergence validity

Primary indicator	Secondary indicator	Specific content	Estimate	AVE	CR
Evaluation of teachers’ digital teaching effect	Satisfaction	Improve the teaching atmosphere	0.8785	0.8515	0.9245
		Improve the teaching activity	0.9055
	Acquired sense	Improving teaching ability	0.9245	0.8545	0.9285
		Improving teachers’ literacy	0.9268
	Student interaction perception	The willingness to interact significantly increased	0.9048	0.8345	0.9125
		The distance between teachers and students is closer	0.8485
Student digital learning effect evaluation	Satisfaction	Promote outlook on life	0.8463	0.8245	0.9154
		Promote values	0.8912
	Acquired sense	Enhanced identity	0.9248	0.8236	0.9048
		Enhanced social identity	0.9078
	Student interaction perception	The willingness to interact significantly increased	0.8626	0.8548	0.9278
		The distance between teachers and students is closer	0.9286
The combination of recommended teaching results	Student evaluation	Student performance progress is obvious	0.8932	0.8315	0.9185
		The comprehensive quality of the students was significantly improved	0.8936
	Teacher evaluation	Teacher satisfaction improvement	0.9185	0.8269	0.9088
		Teachers’ class is high in praise	0.9058

5.2

Descriptive statistics and analysis of grade differences

Table 3 shows the statistical analysis of the effectiveness of teaching Civics and Politics using digital resources.In this study, the data were analyzed using descriptive statistics using SPSS 23.0 software, and the range, extreme value, mean, standard deviation, kurtosis, and skewness of the data were calculated. It was statistically found that the skewness and kurtosis of all the data were between positive and negative [-2,2], indicating that the data were normally distributed. Therefore, an independent samples t-test should be used for the comparison of means. The overall evaluation of Civics teaching by the subject students was high (mean = 79.2988, standard deviation = 8.5489). Among the three dimensions, the level of self-evaluation was the highest (mean = 35.0855, standard deviation = 4.3485), teaching effectiveness was the second highest (mean = 25.3658, standard deviation = 2.8582) while teaching resources and methods scored the lowest (mean = 19.4856, standard deviation = 2.2863). This indicates that students believe that Civic Education recommended by digital resources not only promotes the absorption of specialized knowledge but also effectively improves their ideological and moral qualities.

Table 3.

Descriptive statistics and grade difference analysis

Dimension	Minimum value	Maximum value	Mean value	Standard deviation	Kurtosis	Degree of bias
Self-evaluation	18	42	35.0855	4.3485	-0.2856	0.0587
Teaching effect	16	38	25.3658	2.8582	-0.1836	0.3785
Teaching resources and methods	15	27	19.4856	2.2863	0.3788	0.2678
Overall evaluation	52	94	79.2988	8.5489	-0.6445	-1.2645
Dimension Self-evaluation	T	Freedom	Mean difference(Freshman year-sophomore)	P	95% confidence interval
Dimension Self-evaluation	T	Freedom	Mean difference(Freshman year-sophomore)	P	Lower limit	Upper limit
Teaching effect	1.3452	36.8555	1.4855	0.1826	-0.6885	3.6285
Teaching resources and methods	2.2485	36.8555	0.6785	0.2658	0.0826	1.3885
Overall evaluation	2.1226	36.8555	0.8345	0.1388	0.0548	1.6248
Dimension	1.8455	36.8555	2.9285	0.2658	-0.1236	6.1578

In order to understand whether there is a significant difference in the evaluation of the effectiveness of Civics teaching under digital resource recommendation among students of different grades, this study chose the relevant data of freshmen and sophomores and carried out an independent samples t-test on the data, and even though freshmen students’ evaluation of the Civics teaching both in the whole and in the mean value of each dimension was that of sophomores, the difference was not significant (P>0.05). This shows that both freshmen and sophomores agree with the teaching mode of “digital resources recommendation” and believe that they can improve their ideological and political quality through this teaching mode.

6

Conclusion

This paper focuses on personalized learning and network resources, constructing a personalized network recommendation model for the Civics course resources in colleges and universities, based on the fundamental principles of building Civics education in these institutions. After mining the interests of Civics learners, collaborative filtering technology is used to provide personalized resource recommendations. In order to improve the real-time performance of the college Civics and Political Science resource recommendation model, a hybrid recommendation algorithm is composed by combining collaborative filtering and resource content, and the recommendation effect of the improved model is analyzed according to the evaluation indexes.

Evaluate the model’s recommendation effect of Civics teaching resources in terms of precision and recall, with this paper’s method as the experimental group and the big data recommendation and collaborative filtering algorithms as the control A and control B groups, respectively. The experimental group achieves a precision of matching recommended resources exceeding 89%, and when the parameter K=10 is applied, the recall rate rises to 89.96%, demonstrating a superior recommendation effect in the actual recommendation scenario.

The number of recommendations of this paper’s method ranks first among the three algorithms. For example, the number of recommendations in the control group A is 1295.81448 and 1278.84615, and the number of recommendations in the control group B is 1277.71493. The number of recommendations in the experimental group is 1310.52036 and 1334.27602 for the Civics course resources recommendation of the senior year. This paper’s method effectively integrates different types of resource data to achieve personalized recommendations for civic education resources. Meanwhile, in the favorite degree of recommended courses, the listening time of this paper’s method can be kept above 42 min, and the other two algorithms are only about 20 min.

The overall evaluation of the subject students on the Civic and Political Digital Resources Recommendation Teaching is high, with the mean and standard deviation of 79.2988 and 8.5489, respectively. Independent samples t-test found that the significance p-value of the difference of each dimension is greater than 0.05, and there is no significant difference, which can be seen that all the students agree with this teaching mode, and it can improve their learning literacy.

Funding:

Research on the Construction of Quality Indicator Evaluation System for Ideological and Political Education in Specialized Courses of Higher Vocational Colleges in Shaanxi Province for the 14th Five-Year Plan (2023) Research (Project No. SGH23Q0396).

Langue:: Anglais

Périodicité:: 1 fois par an
Sujets de la revue:: Sciences de la vie, Sciences de la vie, autres, Mathématiques, Mathématiques appliquées, Mathématiques générales, Physique, Physique, autres

RSS Feed de la revue

Research on Digital Resource Construction and Intelligent Recommendation Strategy for Civic and Political Education in Colleges and Universities

Jiayi Xie

Ruihua Zhao

Rui Li

Jing Yang

Publié en ligne: 17 mars 2025

Reçu: 16 oct. 2024

Accepté: 05 févr. 2025

DOI: https://doi.org/10.2478/amns-2025-0169

Mots clésCollaborative Filtering, Preference Similarity, Hybrid Recommendation, Personalised Resources, Civics Courses in Universities

© 2025 Jiayi Xie et al., published by Sciendo

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

Mots clés
Collaborative Filtering, Preference Similarity, Hybrid Recommendation, Personalised Resources, Civics Courses in Universities