Study on the Path of Embedding Red Culture into Civic and Political Education in Colleges and Universities and the Enhancement of Cultivation Effect by Integrating Deep Learning Models

Red culture is rich in revolutionary spirit and heavy historical and cultural connotations, and is a valuable spiritual wealth and cultural heritage of the Chinese nation. In China’s 9.6 million square kilometers of the vast land of red resources scattered, in the party united to lead the Chinese people in the great course of a century of struggle in the red bloodline passed from generation to generation. Therefore, it is necessary to make good use of the red resources, carry forward the red traditions and pass on the red genes [1-4]. Colleges and universities are an important position for training socialist builders and successors for the country, the cradle of talent cultivation, shouldering the important mission of educating people for the Party and talents for the country. Cultivating moral and talented youth is an important goal of talent training in colleges and universities [5-8]. Red culture contains rich nurturing value and value of the times, and embedding red culture into ideological and political education in colleges and universities has important practical significance [9-10]. Through the education of red culture, colleges and universities can let college students understand the history of the struggle of the Communist Party of China (CPC) more deeply, let college students feel the spirit of the revolutionary forefathers’ unremitting struggle to realize national independence, people’s liberation, and the prosperity of the country, so that college students can be more firm in their beliefs in Marxism, beliefs in socialism with Chinese characteristics, and confidence in the great rejuvenation of the Chinese nation, and set up a great ambition of repaying the country and strengthening the country and be a struggling person who can bear the responsibility for the country [11-14]. How to better disseminate the red culture and continue the spirit of revolutionary forefathers has become an important issue for colleges and universities. Using deep learning models to integrate red culture into classroom teaching can cultivate college students’ correct worldview, outlook on life and values, promote the overall development of college students’ morality, intelligence, physical fitness and aesthetics, enhance the nurturing effect of ideological and political education in colleges and universities, and contribute to the great rejuvenation of the Chinese nation [15-19].

The article embeds the deep learning model into the ideological education of colleges and universities, utilizes the deep learning algorithm to recommend the red cultural elements in the ideological content, so as to promote the combination of red culture and the teaching of ideology in colleges and universities, and proposes the path of red culture ideological education based on deep learning. Construct an evaluation index system for ideology and politics education in colleges and universities, and calculate the weight of each index using the hierarchical analysis method. The evaluation index system is used to obtain evaluation results of the red culture ideological and political education path based on deep learning in this paper. Setting up relevant experiments, comparing the changes in the ideological and political literacy of the experimental group and the control group before and after the experiments, so as to analyze the performance of the path of ideological and political education in colleges and universities proposed in this paper in terms of its effect on educating people.

2

Red Culture Civic and Political Education Based on Deep Learning

2.1

Red Culture and Civic and Political Education

2.1.1

Red culture and red cultural resources

1)

Red culture

Intuitively, red culture is an organic fusion of the red symbolism in traditional Chinese culture and the concept of culture [20]. Thus forming a rich and far-reaching red culture. The term “red culture” first originated from scholar Liu Shouli’s academic paper “Study on the Enrichment and Development of the ‘Red Culture’ of the Soviet Union on the Spirit of the Chinese Nation,” in which he argued that red culture refers to red literary and artistic works and related spirits. Although the definition of red culture has not yet been unified in the academic circles, the “red culture” studied in this topic refers to the sum of material and spiritual cultures created by the people during the revolutionary, construction and reform periods to realize the independence and autonomy of the Chinese nation and to move towards a great renaissance. It includes writings, poems, ballads, memorials, museums, red sites, and former residences of red figures.

2)

Red cultural resources

“Red cultural resources” is the carrier of red culture, is the inheritance of the red spirit and red gene of the specific material form [21]. Regarding the definition of the concept of “red cultural resources”, there are currently disputes in the academic community over the time span, the subject of practice and the basic scope: In terms of time span, there is a view that “red cultural resources” are the sum of resources formed during the New Democratic Revolution. Another viewpoint is that “red cultural resources” are the sum of the resources formed during the new democratic revolution, socialist revolution and construction period. The main body of practice: some scholars believe that the Communist Party of China is the main body of “red cultural resources”. Other scholars are of the view that all material and spiritual resources that conform to the trend of the times and contribute to the development of history should be included in the “red cultural resources”. In terms of basic scope, the representative of the category of “red cultural resources” is Zhang Taicheng, who divides “red cultural resources” into tangible and intangible ideological, political, cultural and spiritual products.

In general, cultural resources in red are classified into concrete and non-concrete forms. Specific forms refer to the important carriers for the transmission of the red spirit, such as ruins and relics, memorial sites, historical bibliographies, and literary works, which were formed during the process of revolution and construction. Non-specific forms are the great national spirit and ideal beliefs centered on patriotism, which were gradually formed in the process of revolution, construction, and reform.

2.1.2

Civic education

Ideological and political education refers to the social or social groups with certain ideological concepts, political views and moral norms, to implement purposeful, planned and organized influence on its members, so as to make them form the ideological virtues in line with certain social needs of social practical activities. Carrying out Civic and political education is of great significance to the majority of young students in terms of firming up their ideals and deepening the common ideal of socialism with Chinese characteristics in colleges and universities, and has a significance that cannot be replaced by other courses.

2.2

Recommendation of Red Cultural Resources Based on Deep Learning

With the continuous progress and improvement of deep learning technology, its performance in the field of recommender system has surpassed the traditional recommendation technology [22]. By using embedding layers to represent high-dimensional and sparse features as low-dimensional dense vectors, and then inputting them into neural networks, deep learning can automatically perform feature intersection and mine out hidden variables. This technique not only simplifies the process of feature design and selection, but also improves the accuracy of recommendations.

Through automatic feature crossover, deep learning can capture the complex interactions between users and items, thus better understanding the hidden relationships between users and items. In addition, deep learning can capture more complex patterns of user interests, thus better personalizing recommendations for users. Therefore, deep learning has become a promising technique in the field of recommender systems, which can effectively improve the accuracy of recommendations and user satisfaction.

The recommendation models based on deep learning mainly include Wide & Deep model, Deep FM model, DCN model, XDeepFM model, etc.

In this paper, deep learning is used in the civic education of colleges and universities, and the XDeepFM model is selected for the recommendation of civic red cultural resources [23]. Compared with traditional recommendation algorithms, the XDeepFM model combines explicit feature interaction through the CIN module and implicit feature intersection through the DNN module, eliminating the labor cost of manually mining cross features. Compared with other deep learning based recommendation algorithms, the XDeepFM model inherits the design theory of Wide & Deep model, and connects the CIN module, linear module and DNN module in a parallel way.In addition to combining explicit feature interactions with implicit feature intersections, the XDeepFM model makes the feature interactions occur at the vector level, and adds the linear module to enhance the model memory. The advantages of XDeepFM mainly focus on explicit feature interactions and vector-level feature interactions, and these two important concepts are explained as follows: 1)

Explicit and Implicit Feature Interaction

In automatic feature engineering, there are three directions, which are implicit feature combination, semi-explicit feature combination and explicit feature combination. Among them, although implicit feature combination is very friendly to continuous features, it lacks interpretability, and the lack of interpretability makes it difficult for feature combinations combined by deep learning to be used by other algorithms, and it is also difficult to provide explicit feedback information. Semi-explicit feature combinations are mainly forest class algorithms, which have some interpretability, but still cannot clearly reflect the feature correlation or combination relationship.

Explicit feature combinations will explicitly indicate which features are combined to serve as base features, making feature combinations more interpretable. Explicit feature combination is what is used in XDeepFM.

2)

Element-level feature interaction and vector-level feature interaction

Take two vectors of dimension 3 as an example, the two vectors are $(a_{1}, b_{1}, c_{1})$ and $(a_{2}, b_{2}, c_{2})$ , if the interaction result is $f (w_{1} (a_{1} \cdot a_{2}), w_{2} (b_{1} \cdot b_{2}), w_{3} (c_{1} \cdot c_{2}))$ , then we consider the interaction process occurs at the element level. If the interaction result is $f (w (a_{1} \cdot a_{2}, b_{1} \cdot b_{2}, c_{1} \cdot c_{2}))$ , then we consider that the feature interaction is generated at the vector level.

In order to achieve interaction learning at the vector level and explicitly learn higher-order feature interactions automatically, the XDeepFM model proposes a new crossover network, CIN, in which the neurons of each layer are computed from the hidden layer of the previous layer and the original features, as shown in Equation (1): (1) $X_{h, *}^{k} = \sum_{i = 1}^{H_{k - 1}} \sum_{j = 1}^{m} W_{i j}^{k, h} (X_{i, *}^{k - 1} \circ X_{j, *}^{0})$

Where W_k,h is the parameter matrix of the hrd eigenvector in layer k and ∘ represents the Hadamard product with the formula shown in equation (2): (2) $〈 a_{1}, a_{2}, a_{3} 〉 \circ 〈 b_{1}, b_{2}, b_{3} 〉 = 〈 a_{1} b_{1}, a_{2} b_{2}, a_{3} b_{3} 〉$

The process of generating a vector in layer k is shown in Fig. 1, where X₀ is the original feature input and X_k−1 is the previous hidden layer, and both compute an intermediate result on which the state of the next hidden layer is generated using H_k convolution kernels of size m · H_k−1.

CIN performs explicit interactions between features through vector levels, and the following three formulas can show how CIN performs explicit feature learning. First X₁ the hnd neuron vector expression formula is shown in equation (3): (3) $X_{h}^{1} = \sum_{\begin{matrix} i \in [m] \\ j \in [m] \end{matrix}} W_{i, j}^{1, h} (X_{i}^{0} \circ X_{j}^{0})$

Further, the hnd neuron vector of X₂ can be expressed as shown in equation (4): (4) $\begin{array}{rcl} X_{h}^{2} & = & \sum_{\begin{array}{l} i \in [m] \\ j \in [m] \end{array}} W_{i j}^{2, h} (X_{i}^{l} \circ X_{j}^{0}) \\ = & \sum_{\begin{array}{l} i \in [m] \\ j \in [m] \end{array}} \sum_{\begin{array}{l} l \in [m] \\ k \in [m] \end{array}} W_{i, j}^{2, h} W_{l, k}^{l, i} (X_{j}^{0} \circ X_{k}^{0} \circ X_{l}^{0}) \end{array}$

Finally, the hnd nerve vector of layer k can be represented as shown in equation (5): (5) $\begin{array}{rcl} X_{h}^{k} & = & \sum_{\begin{array}{l} i \in [m] \\ j \in [m] \end{array}} W_{i j}^{k h} (X_{i}^{k - 1} \circ X_{j}^{0}) \\ = & \sum_{\begin{array}{l} i \in [m] \\ j \in [m] \end{array}} \dots \sum_{\begin{array}{l} r \in [m] \\ v \in [m] \end{array}} \sum_{\begin{array}{l} l \in [m] \\ s \in [m] \end{array}} W_{i j}^{k h} \dots W_{l, s}^{l r} (X_{j}^{0} \circ \dots \circ X_{s}^{0} \circ X_{l}^{0}) \end{array}$

With the above three formulas, we can make an explicit representation of the vectors produced at each layer, and can prove that CIN is explicitly learning feature crossover.

CIN is characterized by the fact that the number of layers of the network determines the number of orders of feature interactions to be learned, and each intermediate hidden layer is connected to the output layer through a pooling operation, thus ensuring that the output result contains feature interactions of different orders. Structurally, the structure of CIN is similar to that of recurrent neural networks in that the state of each layer is generated by computing the hidden layer of the previous layer with additional inputs, but unlike RNNs, the additional inputs of CIN are always X₀. The macroscopic architecture of CIN is shown in Fig. 2.

Each hidden layer X_k, $k \in [1, T]$ is associated with the final output and T is the depth of the network . The relationship between X_k and the final output p⁺ is shown in the following equation.

First use summation pooling on the feature maps of each hidden layer: (6) $p_{i}^{k} = \sum_{j = 1}^{D} X_{i j}^{k}, i \in [1, H_{k}]$

This results in a pooling vector p^k, p^k as shown in equation (7): (7) $p^{k} = [p_{1}^{k}, p_{2}^{k}, \dots, p_{H_{k}}^{k}]$

Splice all the pooling vectors to get the output p+, p+ of the CIN module as shown in equation (8): (8) $p^{+} = [p^{1}, p^{2}, \dots, p^{T}]$

By combining the output of the CIN with the linear regression unit and the DNN module, the complete structure of XDeepFM can be obtained, which is shown in Fig. 3.

2.3

The Path of Civic and Political Education in Colleges and Universities Integrating Red Culture

Integrating the red cultural resources recommended by deep learning algorithms into the civic education of colleges and universities is a systematic and fundamental project. It is necessary to integrate the whole project with a macroscopic vision, start from excavating the red cultural resources, explaining and carrying forward the various types of red spirit, set up a team of teachers who can speak well about the red cultural resources, improve the classroom that can transmit the positive energy of the red culture and spirit, and form a group of classes corresponding to the civic education curriculum and the course civic education, the theoretical teaching and the practical teaching. It is also important to develop a scientific evaluation system focusing on cultural resources that are red, so as to achieve the purpose of educating people with culture and moistening their hearts with literature.

2.3.1

Mining red cultural resources to transform them into ideological and political materials

The key to integrating red cultural resources into the construction of “big ideological and political courses” in colleges and universities is to study the connotation of red cultural resources in depth and reveal their contemporary value. First of all, it is necessary to systematically sort out and categorize the red cultural resources, and clarify the characteristics, status quo and application potential of different types of red cultural resources. Relevant data and information can be collected and organized through literature research and field research. In this process, the personal memories and oral information of revolutionary heroes should also be fully explored. In the information age, new media and digital technology facilitate the collection and collation of red cultural resources, and it is necessary to make use of channels such as the Internet, e-books, databases and digital archives to obtain and collate relevant information on red cultural resources.

2.3.2

Optimizing the Civic and Political Education Teaching Staff

It is necessary to expand the faculty of Civic and Political Studies in colleges and universities, offer multiple subjects, emphasize multiple participation options, and establish a system of teacher education for Civic and Political Studies. Colleges and universities should take this opportunity to continuously enrich and expand their faculty and offer multiple subjects of education. To polish the red color, create a team of professional teachers who adhere to the “soul” and “root” of red culture. This professional teaching team should be good at grasping the past, present and future development trend of socialism with Chinese characteristics, guiding young people to confidently and self-reliably shoulder the historical responsibility of national rejuvenation, thickly planting family and national sentiments, and guarding the roots of red history, so that the red genes can be passed on from generation to generation.

2.3.3

Constructing a Civics Classroom Based on Red Cultural Resources

To deepen the construction of the civic and political classroom based on red cultural resources, it is necessary to synchronize the implementation of social practice education while guaranteeing the penetration of red cultural resources into the classroom education of civic and political courses, so as to realize the effective interaction between the two, and enable students to have a strong sense of participation in the red cultural resources, a sense of immersion, and a sense of acquisition.

Deepen the construction of red cultural resources based on the ideological classroom, but also fully utilize the advantages of local red cultural resources, and continuously expand the red cultural education online teaching environment. On this basis, we should be good at applying advanced new media technology to broaden channels, realize information technology, and create a large intelligent classroom.

2.3.4

Creating a cluster of Civics and Political Science courses linked by red cultural resources

To create a group of ideological and political courses with red cultural resources as a link, to realize the deep integration of the ideological and political courses and course ideology, and to improve the synergistic nurturing system of the ideological and political courses and course ideology, it is necessary to break the academic barriers, to promote the reconstruction of the curriculum construction system of the ideological and political courses, it is also necessary to give full play to the roles of various disciplines in the construction of ideological and political courses, to strengthen the coordination of the various courses of the ideological and political courses, to create the golden group of the courses with the characteristics of the red cultural resources, and to fully Tapping the hidden ideological and political education elements contained in each course, especially the red culture nurturing elements.

2.3.5

Building a scientific evaluation system based on red cultural resources

Constructing a pluralistic participation, open and scientific evaluation system can make the evaluation of the integration of red cultural resources into the construction of civic politics in colleges and universities have a clear basis in reality and a specific direction, give full play to the leading role of teachers in teaching activities and the subjectivity of students, and enhance the relevance and effectiveness of the teaching of civic politics courses.

Colleges and universities should implement multiple evaluations combining teacher evaluation, student evaluation and school evaluation, and set up evaluation committees with the participation of teachers, students, industry experts and other parties to ensure the fairness of the evaluation process and make the evaluation system more sound. It is necessary to promote the exchange and collision of diversified opinions, encourage the participation of teachers and students in curriculum design, teaching evaluation and other aspects related to the teaching of red cultural resources, and introduce democratic evaluation and other methods.

3

Empirical analysis of parenting effects

The red culture civic education path based on deep learning proposed in this paper is applied to the civic education of S-school law school, and the nurturing effect of the civic education path in this paper is analyzed through the results of empirical research.

3.1

Evaluation results of Civic Education

3.1.1

Construction of evaluation index system for political education

The author constructs the evaluation index system of civic education and assigns weights to each evaluation index through hierarchical analysis method, and the evaluation index system of civic education and its weights are shown in Table 1.

Table 1.

Ideological and political education

Primary index	Secondary index	Tertiary index
1.Target concept (0.298)	1.1 Course target positioning (0.500)	1.1.1 Training scheme fitness (0.356)
		1.1.2 Target diversification (0.329)
		1.1.3 Education function realization (0.315)
	1.2 Course construction concept (0.500)	1.2.1 The unification of people-education and talent-education (0.524)
	1.2 Course construction concept (0.500)	1.2.2 Good atmosphere creation (0.476)
2.Faculty (0.085)	2.1 Faculty composition (0.425)	2.1.1 Teacher’s advanced ideological and political consciousness (0.526)
	2.1 Faculty composition (0.425)	2.1.2 Abundant teachers and stable personnel (0.474)
	2.2 Teacher’s ideological and political literacy (0.575)	2.2.1 The right political position and belief (0.602)
	2.2 Teacher’s ideological and political literacy (0.575)	2.2.2 Good at excavating ideological and political elements (0.398)
3.Course resource (0.116)	3.1 Red cultural resource application (0.478)	3.1.1 Abstract red cultural resource (0.500)
	3.1 Red cultural resource application (0.478)	3.1.2 Specific red cultural resource (0.500)
	3.2 Ideological and political material repository (0.522)	3.2.1 Rich and systematic content (0.411)
	3.2 Ideological and political material repository (0.522)	3.2.2 Timely-updated and convenient repository (0.589)
4.Teaching implementation (0.322)	4.1 Teaching arrangement (0.242)	4.1.1 The course theory and the practice are reasonable (0.389)
	4.1 Teaching arrangement (0.242)	4.1.2 The teaching progress is reasonable and rigorous (0.611)
	4.2 Teaching content (0.263)	4.2.1 Emphasis on combination of theory and practice (0.375)
		4.2.2 Focus on red culture resource (0.352)
		4.2.3 The proportion of ideological and political material is reasonable (0.273)
	4.3 Teaching method (0.257)	4.3.1 Enrich the present form of ideological and political content (0.195)
		4.3.2 Strengthen experience of the students (0.207)
		4.3.3 The teaching method is flexible and appropriate (0.213)
		4.3.4 Focus on inspiring students’ thinking (0.182)
		4.3.5 Timely, appropriate use of teaching tools and modern technical means (0.203)
	4.4 Course assessment (0.238)	4.4.1 Process evaluation is combined with the final assessment (0.342)
		4.4.2 Diversified assessment subject (0.328)
		4.4.3 Diversified assessment form (0.330)
5.Teaching effect (0.179)	5.1 Learning behavior (0.415)	5.1.1 High class attendance rate (0.228)
		5.1.2 Good class condition (0.284)
		5.1.3 Active class discussion (0.267)
		5.1.4 High homework-completion rate (0.221)
	5.2 Learning achievement (0.585)	5.2.1 Proficient in mastering and applying the course theory (0.372)
		5.2.2 Exercise and strengthen comprehensive ability (0.309)
		5.2.3 Learning attitude is more positive (0.319)

3.1.2

Analysis of evaluation results

1)

Reliability and validity test

264 questionnaires on Civic Teaching were distributed, 264 were recovered, with a recovery rate of 100%, and 260 valid questionnaires, with a validity rate of 98.48%. The questionnaire was tested for reliability and validity using Cronbach’s reliability coefficient (Cronbach’ α coefficient value) and KMO value. The results of the reliability test are shown in Table 2.

As can be seen from Table 2, the Cronbach’ α coefficient of the questionnaire is 0.956, which is greater than 0.7, indicating that the indicator system applies the high internal consistency of the retrieved data, which passes the reliability test and is suitable for use. Validated by the factor analysis method, the analysis yielded a KMO value of 0.952, greater than 0.7, indicating that the structural validity of the data retrieved by the application of the indicator system is better, i.e., the validity of the questionnaire is higher.

2)

Evaluation results

The results of evaluating the deep learning-based red culture civic teaching with the civic teaching questionnaire are shown in Figure 4. All the indicators are scored out of 5. As can be seen in Figure 4, the average score of each indicator is between 4 and 5, and the standard deviation is less than 1. Among the first-level indicators, the overall score of the teaching implementation dimension is the highest (4.67), and the following ones are, in order of priority, the teaching effect (4.55), the teaching faculty (4.50), the target concept (4.49), and the teaching implementation (4.28). Among the secondary indicators, the highest scoring indicator is teaching content with 4.77 points, and the lowest scoring indicator is the Civics Material Library with 4.14 points. The scores of the three-level indicators are concentrated in the range of 4.07~4.91, with the highest scoring indicator being “rational and rigorous teaching process” and the lowest scoring indicator being “enriching the expression of civic and political content”. It can be seen that the overall evaluation results of the path of ideological education in this article are better, with an overall score of 4.54.

Table 2.

Reliability and validity of the questionnaire

Cronbach’ α coefficient		Item number
0.956		33
KMO value		0.952
Bartlett	Approximate chi-square	11054.5516
	df	385.000
	p	0.000

3.2

Effectiveness in educating people

A class in the second year of the Law School of School S was randomly selected as the experimental group of the parenting effect experiment, and then a class in the second year of the College of Arts and Letters of the same school (which adopts the traditional Civics and Politics teaching mode) was randomly selected as the control group. By comparing the two classes’ semester-long Civics literacy (classroom theory, practical activities, red culture papers, group research, and Civics grades), we will explore whether this paper’s deep-learning-based Civics education path in colleges and universities is effective.

3.2.1

Comparative results of Civic Literacy of the two groups before and after the experiment

In order to be able to better observe and compare the effects of the two teaching modes, tests were conducted on the experimental and control groups before and after the start of the experiment. The significance analysis of the experimental and control groups’ Civic and Political Literacy was carried out on the students of the two classes. The results of the analysis are shown in Table 3.

Table 3.

Comparison of pre-test ideological and political literacy of two groups

Item	Pre/post-test	Experimental group	Control group	t	p
Class theory	Pre-test	13.56±2.95	13.84±2.58	-0.265	0.624
Class theory	Post-test	18.96±4.87	13.92±2.94	5.548	0.003
Practice activity	Pre-test	12.68±3.11	13.56±3.06	0.523	0.428
Practice activity	Post-test	19.07±5.06	14.05±3.46	6.845	0.002
Red culture thesis	Pre-test	13.27±2.64	13.08±2.77	0.842	0.578
Red culture thesis	Post-test	19.53±4.93	13.66±3.78	6.418	0.001
Team research	Pre-test	13.74±2.59	12.89±2.91	0.642	0.689
Team research	Post-test	19.88±5.27	13.26±3.09	5.915	0.001
Ideological and political grade	Pre-test	12.73±2.82	13.02±3.22	-0.459	0.816
Ideological and political grade	Post-test	19.42±5.42	13.68±3.48	6.942	0.003

As can be seen from Table 3, the p-value results of the indicators related to Civic and Political Literacy (classroom theory, practical activities, red culture paper, group research, and Civic and Political achievement) of the experimental and control groups before the experiment were 0.624, 0.428, 0.578, 0.689, and 0.816 (p > 0.05). The results of the experiment show that there is no significant difference between the Civics literacy of the two groups before the experiment, and it is possible to exclude the factors that affect the smooth implementation of the experiment due to the large difference in the level of Civics literacy between the two groups.

After the experiment, the results show that the scores of the experimental group’s Civic and Political Literacy related indexes are higher than those of the control group, and the score difference reaches 5.04, 5.02, 5.87, 6.62, and 5.74 respectively.By analyzing the results of the test on the level of Civic and Political Literacy of the experimental group and the control group, it is concluded that the p-values are 0.003, 0.002, 0.001, 0.001, 0.003 respectively (p<0.05), and there is a significant differences.

In summary, in the teaching of Civics in colleges and universities, the effect of the red culture Civics education path based on deep learning is more obvious in improving students’ Civics literacy compared with the traditional Civics teaching mode.

3.2.2

Pre and Post Comparison Results of Civic Literacy in the Two Groups

After adopting the traditional Civics teaching mode for one semester in the control group, in order to conduct a more in-depth analysis of the impact of the Civics education path proposed in this paper and the traditional Civics teaching mode on the level of students’ Civics literacy in Civics teaching for the students in the experimental group and the control group before and after the experimental group, respectively, we did a paired-sample t-test, and the test results of the experimental group are shown in Table 4, and the results of the control group are shown in Table 5.

Table 4.

Pre-test and post-test ideological and political literacy comparison of experimental group

Item	Pre-test	Post-test	t	p
Class theory	13.56±2.95	18.96±4.87	-4.551	0.004
Practice activity	12.68±3.11	19.07±5.06	-8.128	0.001
Red culture thesis	13.27±2.64	19.53±4.93	-7.105	0.002
Team research	13.74±2.59	19.88±5.27	-8.548	0.001
Ideological and political grade	12.73±2.82	19.42±5.42	-8.574	0.001

Table 5.

Pre-test and post-test ideological and political literacy comparison of control group

Item	Pre-test	Post-test	t	p
Class theory	13.84±2.58	13.92±2.94	-0.484	0.548
Practice activity	13.56±3.06	14.05±3.46	-0.566	0.624
Red culture thesis	13.08±2.77	13.66±3.78	-0.586	0.678
Team research	12.89±2.91	13.26±3.09	-0.879	0.795
Ideological and political grade	13.02±3.22	13.68±3.48	-0.785	0.852

As can be seen from Table 4, after the experiment, the experimental group’s performance in the indicators related to Civic and Political Literacy (classroom theory, practical activities, red culture paper, group research, and Civic and Political Achievement) was significantly improved, and the dimensions got 5.40, 5.39, 6.26, 6.14, and 6.69 scores, respectively, with p-values of <0.05, which is a significant difference. It shows that adopting the red culture political education path based on deep learning proposed in this paper in the teaching of political thinking in colleges and universities can significantly improve the level of students’ political literacy.

As can be seen from Table 5, after the control group adopts the traditional Civics teaching mode for teaching, the scores of Civics literacy related indexes are not significantly improved, and the p-values are all > 0.05, with no significant difference. It shows that adopting traditional teaching modes in the teaching of Civics in colleges and universities does not have a significant effect on improving the level of students’ Civics literacy.

4

Conclusion

The article combines deep learning algorithms with ideological education in colleges and universities, utilizes deep learning to recommend red cultural resources, promotes the integration of red culture and ideological education, and thus puts forward the path of red culture ideological education based on deep learning. The evaluation results of the ideological education path of this article and its nurturing effect are obtained through empirical research. 1)

The path of red culture ideological and political education based on in-depth learning in this paper scores more than 4 points in each index. The first-level indicators in the order of scores from highest to lowest are: teaching implementation (4.67), teaching effect (4.55), faculty (4.50), target concept (4.49), and teaching implementation (4.28). The highest-scoring indicator in the secondary indicators is teaching content (4.77), and the lowest-scoring indicator is the Civics Material Library (4.14 points). The score range of tertiary indicators is [4.07, 4.91], and the overall score is 4.54. The path of Civics education proposed in this paper has achieved high evaluation results.

2)

The p-value of the Civic and Political Literacy related indexes of the experimental group and the control group before the experiment is greater than 0.05. There is no significant difference between the Civic and Political Literacy of the two groups. After the experiment, the scores of the Civic and Political Literacy related indexes of the experimental group were higher than those of the control group by 5.04, 5.02, 5.87, 6.62, and 5.74 points respectively. The p-value for each dimension is less than 0.05, and the differences between the two groups are significant. The experimental group’s scores on the indicators related to Civic and Political Literacy were significantly improved, and all dimensions got more than 5 points, with p-values less than 0.05. The scores on the indicators related to Civic and Political Literacy of the control group were not significantly improved, and the difference between the before and after scores of all dimensions was not more than 1 point, with p-values greater than 0.05.

Język:: Angielski

Częstotliwość wydawania:: 1 razy w roku
Dziedziny czasopisma:: Nauki biologiczne, Nauki biologiczne, inne, Matematyka, Matematyka stosowana, Matematyka ogólna, Fizyka, Fizyka, inne

Kanał RSS czasopisma

Study on the Path of Embedding Red Culture into Civic and Political Education in Colleges and Universities and the Enhancement of Cultivation Effect by Integrating Deep Learning Models

Dan Li

Pei Zhao

Data publikacji: 24 mar 2025

Otrzymano: 23 paź 2024

Przyjęty: 15 lut 2025

DOI: https://doi.org/10.2478/amns-2025-0701

Słowa kluczoweDeep learning, Recommendation technology, Red culture, Civic and political education, Nurturing effect

© 2025 Dan Li et al., published by Sciendo

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

Słowa kluczowe
Deep learning, Recommendation technology, Red culture, Civic and political education, Nurturing effect