Abstract
The public’s desire to protect their own interests has increased in recent years. As information technology advances, the number of Internet users grows, the penetration rate improves, and the number of self-media users of network new media grows. Using historical data to predict future behavior and learn from massive amounts of data, humanity has entered the new media age, thanks to the rapid advancement of mobile Internet technology. It solves the problem that traditional database management systems cannot find hidden relationships and rules in data and cannot predict the future based on the data they have. This paper investigates the use of DM technology to intelligently govern network public opinion. Finance, telecommunications, network-related industries, retailers, manufacturing, medical care, pharmaceutical industry, and scientific fields have all made extensive use of it. Data mining (abbreviated as DM) is a type of data analysis that extracts credible, novel, effective, and understandable patterns from large amounts of data. It is closely related to database knowledge discovery.
1. Introduction
In recent years, the public’s desire for their own interests has been increasing. Coupled with the continuous development of information technology, the number of Internet users has been increasing, the popularity rate has been further improved, and the number of online new media users has been increasing. The network has increasingly become an important way for the public to express their views and participate in social management. With the continuous development of mobile internet technology, human beings have entered the New media age [1, 2]. In the new media age, great changes have taken place in the way, object, and content of popular opinion management, which makes it more difficult. Therefore, the governance ability of online popular opinion has become an important aspect that affects the government’s ruling ability. If the governance effect of online popular opinion is not good, it will damage the government’s image and credibility, even because of the negative popular opinion, and it will lead to the intensification of social contradictions and affect social harmony and stability. The development of new media has made great and profound changes in the way of information dissemination and popular opinion in the whole society, and online popular opinion has evolved into a new expression of popular opinion and social force [3, 4]. It can be seen from the hot online popular opinion events in 2018 that online popular opinion has become the main channel for the public to express popular opinion, and there are many popular opinion events related to people’s livelihood issues, which shows that there are still many social contradictions at present [5]. Study the issues that exist in network popular opinion management in the new media age, constantly explore the best countermeasures for popular opinion management, actively respond to the challenges of popular opinion control in the new media age, and achieve accurate popular opinion management in order to promote the continuous development of a harmonious society. In this context, scientific research on popular opinion, proper popular opinion guidance, and network popular opinion governance mode innovation will aid in lowering the cost of public political participation, improving the quality of public services, and establishing a clear, stable, harmonious, and safe network popular opinion space [6, 7].
Then, DM technology is used to mine the data to form the corresponding statistical information, which is convenient for query and output. DM is defined as the popular opinion collection module in the new media age is an important basic guarantee for network popular opinion analysis [8, 9]. Netizens hope to find valuable information for themselves in the shortest time, and the personalized service customized by the department can fundamentally solve this problem. Today, when the data production and transmission capacity is far greater than the data analysis capacity, people hope to provide a higher-level data analysis function to automatically and intelligently convert the data to be processed into useful information and knowledge [10, 11]. It refers to the process of collecting web page information in major web sites and establishing the original web page database by using search engines and then uniformly storing the collected information and data to the network storage system. The research fields of DM have covered industries including finance, telecommunications, network-related industries, retailers, manufacturing, medical care, pharmaceutical industry, and scientific fields [12, 13]. DM is a data analysis technology that extracts credible, novel, effective, and understandable patterns from a large number of data and is closely related to knowledge discovery in database.
Popular opinion on the internet and popular opinion on social media interact and influence one another. In terms of content expression, internet and social popular opinion are similar. To some extent, the development trend of social popular opinion will be influenced by internet popular opinion. As a result, using DM technology to study Internet popular opinion is extremely innovative [3]. It is difficult to dig deep into the deeper information features of the sudden popular opinion crisis in reality by studying the effective intervention of popular opinion crisis through information management without the support of a strong technical background. As a result, the purpose of this paper is to investigate the use of DM technology in social network popular opinion crisis management, which will be crucial in improving social crisis management capabilities and developing a future popular opinion crisis response plan [14]. When we compare all of the knowledge we need and the information we want to get to a large database, DM technology is a critical technology that allows you to quickly find the information and knowledge you need. It is now of great importance to the Internet, which has a large database, and it is a necessary technology for a large number of users.
2. Related Work
In order to improve the ability of social crisis management, literature [15] proposes three paths for the government to dredge popular opinion on the Internet in emergency situations: improving the government’s monitoring and guidance of popular opinion, improving the news release mechanism, and immediately investigatingm and punishing Internet rumors according to law. We can learn from keyword statistics that there are many online popular opinion articles on communication, journalism, pedagogy, sociology, and politics, according to literature [16]. At the same time, it reflects the need for interdisciplinary support in popular opinion research. A statistical comparison of document classification and navigation categories can also help to confirm the keyword statistics comparison results. According to literature [17] research, NSFC’s support for network popular opinion and related fields was also concentrated in recent 2007 and 2008, distributed under the discipline categories of journalism and communication, library and information science, and the “general project funded in 2007” research on the new pattern and mechanism of popular opinion guidance under the new situation, “and the youth projects” research, and implementation of active monitoring sytems.” According to the literature [18], for different life cycle stages of network popular opinion and knowledge integration in each stage, a knowledge integration framework based on life cycle theory should be established. Literature [19] constructed the network popular opinion situation evaluation index system through the big data analysis method, and put forward the quantitative method of three-level indicators in the government negative network popular opinion situation evaluation index system. Literature [20] research shows that the characteristics of Internet universality, immediacy, openness, sharing and interaction and rich, colorful, convenient, and practical application forms determine that it has increasingly become an important position to reflect social conditions and popular opinion. Online hot spots emerge one after another, and the influence of network popular opinion on national affairs and public affairs decision-making is also increasing. Literature [21] points out that most articles on the research of online popular opinion focus on the understanding of online popular opinion from the cultural point of view, focusing on popular opinion, social mentality, and network. The main challenges brought by network popular opinion communication to Chinese cultural security, the order of political life, and the influence of social stability are closely related to the network response strategies of events and so on. According to big data analysis [22], the current state of research on Internet popular opinion in China is still weak, and the literature materials are insufficient. Despite the fact that it has attracted the attention of some experts and scholars from various fields, it is uncommon to conduct in-depth research into its mechanism and related technical support [23]. Literature according to research, China’s online popular opinion research is still in its infancy. Although a group of scholars has entered this new field to conduct related research, there are still a number of issues: in terms of the total amount of related literature, development, and changes, research on online popular opinion has only recently begun and is on the rise, the research level in this field needs to be improved, and related fields must be combined to form large-scale popular opinion research. According to literature [24], online popular opinion is the sum of various emotions generated by the public via the network, which primarily includes attitude, emotion, cognition, and behavior tendency, among other things.
This paper studies the intelligent governance of network popular opinion based on DM technology in the new media age. In the field of network popular opinion, the research of its basic theory has been relatively mature, but there is still a large research space in the network popular opinion crisis and network popular opinion governance, especially with the advent of the new media age, in the context of new media. The research on the governance of network popular opinion is still in a relatively blank state, and this research perspective will become a realistic hot issue. The new media has greatly improved the efficiency of information dissemination, and the information broke out and became extremely popular. With the rapid spread of information, the outbreak and spread of all kinds of information are almost synchronous, which is difficult to distinguish effectively, which poses a great challenge to the management of popular opinion.
3. Principle and Model of DM
Under the current trend of social development led by informatization, technological innovation has reached a new peak, including Internet, big data, and artificial intelligence, which has further facilitated people’s production and life and produced great application value. The response of functional departments to the network popular opinion crisis needs to be made transparent to Internet users on a special system platform. With the support of a large number of other professional technologies, the ability of collecting, analyzing, processing, processing, and storing information can be improved, the service scope of popular opinion research can be broadened, the rhythm of popular opinion and online popular opinion information research can be accelerated, the period from popular opinion collection to generation and publication of popular opinion research results can be shortened, so as to improve the timeliness of popular opinion research and the corresponding speed of popular opinion service, and the accuracy and credibility of popular opinion research results can be improved, so as to meet the needs of various users in the information society.
At present, the research on network popular opinion governance in the New media age is still in its infancy, but practice needs the guidance of theory. Therefore, the research on network popular opinion governance of local governments in the new media age has high theoretical and practical significance. Particle swarm optimization is used to optimize the parameters of BP neural network. Let , , …, …, be input vectors of BP neural network, , , …, be output values, ωij and ωjk be weights, and a typical topology diagram of BP neural network is shown in Figure 1.
Among many grey models, the single sequence first-order linear differential equation model GM (1, 1) model in the grey system is the most commonly used.
The original data sequence is set, and is the number of data.
where .
The corresponding differential equation model is established:
where is the development coefficient, and is the grey action.
Using accumulation generated data to construct accumulation matrix and constant term vector YN, that is,
Solving grey parameters by the least square method,
The solution of the differential equation is
Progressive reduction is as follows:
Let the purity of any category cluster formed after network popular opinion data clustering be defined as
where is the number of documents belonging to the predefined class I and assigned to the -th cluster; NR is the number of documents in the -th cluster category.
As an important achievement of modern scientific and technological innovation, the emergence of DM provides conditions for automatically and intelligently transforming the massive data on the Internet into useful information and knowledge. Through the combination of system automation and manual intervention, through a definable processing process, it can be used as an important channel to understand all kinds of information and a tool for decision support, so as to improve the rapid response ability. Whether from the architecture or specific methods, the DM algorithm can be well integrated into the popular opinion information system. Generally speaking, DM is a circular process. The outbreak of network popular opinion is often caused by the asymmetry of information. Especially in the era of knowledge economy, the trend of cross penetration among disciplines is obvious, the total amount of information increases sharply, and the update speed is accelerated. People’s demand orientation has developed from simple acquisition to personalization and specialization, and network popular opinion has become more personalized and specialized. DM technology is application-oriented from the beginning. At present, DM can play an important role in many important fields, especially in commercial applications such as banking, telecommunications, insurance, transportation, and retail.
Several knowledge bases should be established during the system development process, and technologies like vertical search, knowledge discovery, automatic word segmentation and extraction, automatic classification and clustering, and content analysis should be used to monitor massive network data and automatically discover and analyze network popular opinion. The core of the system’s design is to gather sufficient data on the theme, sort it into standard data, mine it, and output useful decision data. The system structure diagram is shown in Figure 2.
The premise of network popular opinion analysis is to obtain enough and complete network data based on a topic. In order to solve this problem, it is considered to establish data collection rules to achieve this purpose during system construction. It is necessary to convert the collected data into useful information. After collecting relevant data, format, clean, and process the collected data in advance and finally form standardized and suitable statistical data. Then, DM technology is used to mine the data to form the corresponding statistical information, which is convenient for query and output, in order to preprocess the data before DM analysis and form the final effective data to be analyzed. Then, conduct data cleaning to “wash away” the “dirty.” Alert the matched content and generate alert records. At the same time, support the alarm mode of mobile phone and e-mail and send relevant information to preset personnel in time. The module includes keyword matching component and alarm processing component. The preprocessed data is only data and does not form useful information. Therefore, the DM analysis and processing steps are introduced to deeply mine and analyze the data by using the mining technologies such as popular opinion behavior pattern analysis and content analysis, so as to form the core information data of the system and provide a data source for data display.
The network popular opinion collection module, the network popular opinion storage module, the network popular opinion analysis module, the network popular opinion retrieval module, and the network popular opinion publishing module, which together form the intelligent management and guidance platform for network popular opinion in the new media age, can be divided into five functional modules. The design framework of intelligent management and guidance platform for online popular opinion in the new media age is shown in Figure 3.
The popular opinion collection module in the new media age is an important basic guarantee for network popular opinion analysis. It refers to the process of collecting web page information in major web sites and establishing the original web page database by using search engines and then uniformly storing the collected information and data to the network storage system. The characteristics of network popular opinion in the new media age require relevant government departments to quickly take countermeasures, actively guide the transformation of network popular opinion to a positive direction, and maintain social harmony and stability. At present, the research on network popular opinion governance in the new media age is still in its infancy, but practice needs the guidance of theory. Therefore, the research on network popular opinion governance of local governments in the new media age has high theoretical and practical significance. Finally, through the analysis of popular opinion analysis system and a series of retrieval of retrieval system in the new media age, as well as the information and data in the storage system can be sent and displayed to users through the popular opinion reporting system, this is the whole workflow of the simple popular opinion intelligent governance and guidance platform in the new media age.
4. Thoughts on the Construction of Intelligent Governance System of Internet Popular Opinion in the New Media Age
4.1. Intelligent Governance System of Network Popular Opinion Based on DM Technology in New Media Age
Internet popular opinion is the concrete expression of the general public’s demands on the emerging carrier of the Internet, which reflects the public’s views on the needs of public services and social problems. The analysis of popular online popular opinion based on DM technology can help people find out the internal influencing factors of its generation and change from the miscellaneous data information and behavior phenomena and finally get the relationship between the changing rules of popular online popular opinion and influencing factors, which has important practical significance for its deeper guidance and disposal. In the new media age, public mood, attitude, cognition, and behavior can be measured more accurately with the help of big data analysis, which makes the potential expression become apparent. In the new media age, we should make good use of the openness and sharing characteristics of big data, change “you ask me to answer” to “ask for advice from the people,” improve the issue setting, popular opinion mobilization, and public service capabilities of government departments through online interactive platforms such as online politics and government affairs disclosure, establish a smooth and effective dialogue mechanism between the government and the people, and continuously improve the decision-making level and administrative efficiency in the new media age. In the new media environment, with the increasing desire of the public to express their opinions, the traditional passive solution mode can no longer meet the current needs, and the role of social managers has changed to that of public service providers. This requires changing the basic concept of network popular opinion governance, breaking away from traditional management thinking and shaping service-oriented thinking based on the actual needs of the general public. In the face of the complicated and changeable network popular popular opinion, DM technology can divide all records into different classes as much as possible without knowing how many classes there are in the target database in advance and minimize within the same cluster and maximize within different clusters based on a certain measure of similarity, so as to establish the relationship between data attributes. DM has the following six different functions: association analysis, time series pattern, classification, clustering, prediction, deviation analysis, etc. It can also realize that mine of complex data types.
These powerful functions are extremely useful in network popular opinion research. They can conduct targeted mining and analysis of public opinion data, accurately study and judge the network’s current public opinion dynamics, respond quickly to the network’s hot spots, focus, and sensitive topics, and determine the best time to deal with crisis events. In the new media age, the spread of network popular opinion is significantly aided. If the government continues to be the primary body responsible for network popular opinion governance, the government will be ineffective in governing network popular opinion events due to the spread of network popular opinion and the government’s own limitations. Classification analysis is one of the main tasks of DM: it identifies the connotation description of a category, which represents the overall information of this type of data and distinguishes it from other data, in order to construct rules or a decision tree model. It is currently in an advanced stage of development. This governance model aims to fully absorb the multiforces of network popular opinion events, establish a cooperative relationship based on mutual trust, jointly create a good network ecological environment, and achieve the synergistic effect of jointly governing the network society through regular communication and coordination, full resource integration, information sharing, and crisis co governance in the event of an outbreak. We can preliminarily screen a large number of network information according to our own needs, classify, and group all types of popular opinion information, such as setting up “people’s livelihood issues,” “emergencies,” “public security,” and “economic crisis,”, in order to make preliminary preparations for the next work in network popular opinion research.
4.2. Experimental Results and Analysis
The emergence of new media has had a great impact on the government’s governance of online popular opinion. The characteristics of network popular opinion in the new media age require relevant government departments to quickly take countermeasures, actively guide the transformation of network popular opinion to a positive direction, and maintain social harmony and stability. In the new media age, we should strengthen the supervision of online information, improve the accountability mechanism; strengthen the guidance check and content audit of self-media; maintain a high-pressure situation against harmful information such as online violence, online rumors, and historical nihilism; reduce the breeding and dissemination space of illegal and negative information; and create a clean and positive online popular opinion environment. In the new media age, service-oriented government is a communication channel that adapts to changing needs, uses modern information technology to reestablish and reestablish relevant functional organizations, realizes the reengineering of organizational structures and working procedures, creates an intelligent government network service system, and smooths the two-way interaction between the government and the public. The public’s opinions on the Internet are presented on the Internet. To reach the vast number of netizens, new media relies on emerging technology tools. Public opinion, new media, and the Internet are all intertwined. The situation of online popular opinion has changed throughout the communication process. To understand the current situation of online popular opinion in the new media age and how to optimize the construction, we must first understand the current situation of online popular opinion in the new media age and then conduct three experiments. By gathering and sorting data from relevant public opinion research institutions, Chinese netizens primarily use mobile phones to access the Internet, with the highest proportion of 98.7% in 2015-2021, while the proportion of people using desktop computers to access the Internet continues to decline, with only 46% in 2018, down 5 percentage points from 2017. Figures 4–6 show the results of three experiments.
The results of the three experiments show that the first experiment reveals that Chinese Internet users primarily use their mobile phones to access the Internet. In the first experiment, the percentage of people who used mobile phones to surf the Internet reached 65.5 percent in 2016, and the percentage of people who used desktop computers to surf the Internet fell by four percentage points from 2016 to 2017, to 30.45 percent. The second experiment revealed that Chinese Internet users primarily use their mobile phones to access the Internet. In the second experiment, the percentage of people who use mobile phones to surf the Internet reached 58.7% in 2020, while the percentage of people who use desktop computers to surf the Internet fell to 32.45% in 2021, down 4 percentage points from 2019. According to the findings of the third experiment, Chinese Internet users primarily use their mobile phones to access the Internet. In the third experiment, the percentage of people who use mobile phones to surf the Internet reached 53.6 percent in 2020, while the percentage of people who use desktop computers to surf the Internet fell by 6 percentage points from 2016 to 2017, to only 23.68 percent.
At present, the frequency of Chinese Internet users accessing the Internet in fixed places has decreased, which indicates that the public’s dependence on fixed Internet devices is decreasing, and more mobile devices are used to access the Internet. Compared with 2017, the proportion of Internet access at home continues to decrease, currently 81.1%, while the proportion of Internet access at work, schools, and public places has increased. Four experiments were conducted for comparison, as shown in Figures 7–10.
The experimental results show that through the analysis of Figures 7–10, the overall frequency of Chinese Internet users accessing the Internet in fixed places has decreased, indicating that the public’s dependence on fixed Internet access devices for using the Internet has been decreasing, and more mobile devices are used to access the Internet. The proportion of accessing the Internet at home continues to decline compared with 2020-2021, with a minimum reduction of 33%, while in units. The proportion of Internet access in schools and public places has increased. Users will track specific events or popular opinions. To solve this problem, the data related to the object is specially analyzed and processed in the analysis process to form the object’s information chain and store it in the database, using the system’s object tracking component and the characteristics of the object that requires special tracking set by users in the system. You can use charts and other methods to display relevant information from the object tracking chain during data presentation, using historical data to forecast future behavior and extract knowledge from large amounts of data. It overcomes the flaw that a traditional database management system cannot find hidden relationships and rules in data or forecast the future based on current data. The emergence of DM creates the conditions for automatically and intelligently transforming massive amounts of data from the Internet into useful knowledge and information. The asymmetry of information frequently causes the spread of network popular opinion. On a special system platform, the response of functional departments to the network popular opinion crisis must be made transparent to Internet users. Netizens hope to find useful information in the shortest time possible, and the Department’s personalized service can fundamentally solve this problem.
5. Conclusions
The introduction and use of the intelligent monitoring system not only improves the current network public opinion environment in the new media era but it also provides more reliable and correct network public opinion in the new media era for more social organizations, businesses, and governments. With the rapid advancement of information technology, data mining technology’s application fields are becoming increasingly diverse. It provides a set of guarantees for resolving social problems and inconsistencies more effectively. We can maximize the benefits of digitalization by using data mining technology to dig and analyze data on the public opinion crisis in the new media era in depth and then improve the entire society’s ability to cope with the network public opinion crisis in the new media era. The study of network public opinion in the new media era is beneficial to protecting the overwhelming majority of people’s fundamental rights, establishing an efficient and clean government, promoting the development of a harmonious society, and promoting social fairness and justice. These powerful functions are extremely useful in network public opinion research. They can dig and analyze public opinion data in a targeted manner, accurately judge current network public opinion trends, respond quickly to network hot spots, focuses, and sensitive topics, and seize the best opportunity to deal with crisis events.
Data Availability
The data used to support the findings of this study are included within the article.
Conflicts of Interest
The author does not have any possible conflicts of interest.
Acknowledgments
The study was supported by “Scientific Research Fund Project of Xijing University in 2020—study on the effect of public opinion release by new government media in crisis communication, (Grant No. XJ200110).”