Handling of college students’ network public opinion crisis based on big data analysis

Abstract

As an important public utility organization and academic research institute, Colleges feature high density people, high social attention, and relatively free atmosphere of interchange ideas, where network opinion crisis are frequent. Therefore, it is critical for colleges to concentrate more on the control and management of the network opinion crisis. In the paper, it compares events of college students’ online public opinion crisis with other references. Social online public opinion crisis solutions is use for reference. Aiming at the peculiarities of college students’ mental health, it is concluded the research coping strategies based on innovative and comprehensive use of big data technology. We establish a complete emergency protection system to effectively deal with the mental health problems of college students.

Keywords

College online public opinion crisis management big data

1. Exploring of the laws of college students’ cyber public opinion crisis

Recent years, events about online public opinion associated with college students have been on a rise, such as “A Girl kicked by A Female College Student” in 2017, “the Misleading Cadre Appointment Announcement in Sun Yat-sen University” in 2018, and etc. In 2020, such events witnessed an year-on-year increase of 33%, up to 133% from 2018, according to statistics from the database of People’s Net Public Opinion Data Center.

Figure 1.

National serious network public opinion events in colleges during 2018–2020.

How to respond to public opinion crisis scientifically has become an essential subject. For colleges, the settlement of this kind of crisis is more complicated, involving network communication, college students’ psychological development, ideological and political work of colleges, public management, crisis management, and so forth [1]. If disposed wrongly, it is very likely to cause mass incidents, which will pose a severe impact on the physical and mental health of students and the reputation of schools.

Based on the causes of public opinion events, that is college students’ reaction and participation in the current public opinion events, the public opinion is basically classified into the following types:

Interest-oriented type: network widely concerned events that are expressed online by college students who fail to receive effective responses of their reasonable requests on campus [2].

Conflict-related type: events that are broadly spread in the Internet, usually because of the personal divergences and conflicts between teachers and students or students and students not being resolved in time.

Emotionally cathartic type: caused by inappropriate attitudes, situation, and words that are used by college students to express personal emotions in the cyber world [3]. For instance, the “Jie Jieliang Event”, triggered by the fault remarks of a postgraduate student in Xiamen University in April, 2018, leads to an uproar.

Moral judgement type: behaviors, thoughts, and attitudes of individual or groups of college students are amplified by the Internet, bringing a huge concern under the name of morality of netizens [4]. For example, in June 2018, a female college student in Changsha was exposed and even searched by netizens for posting life details with her black boyfriend more often.

Key laws of network public opinion events regarding college students:

Rapid speed to sprawl and heat discussion. College students usually obtain a higher social attention on the account of their young age, high degree, and broad participation in the Internet.

An earlier intervention of colleges and universities can better boost the guidance effectiveness. According to the implicated rules of authority effect in psychology, if colleges can intervene earlier as an official department, the authority will be strengthened in some extent, which promotes the events to reach a positively suggestive effect [5]. Thus, the interventions of schools can not only alleviate public mood to build up a mass trust basis for subsequent handling, but also ensure an orderly disposal through an effective instruction and coordination of different divisions.

The transmission falls with a wave-like manner. Seeing from the rules of transmission, public’s attention to the events, particularly negative ones, more or less presents a waving decrease from the beginning “extreme excitement” to the final “indifference” [6]. That indicates the events are far more concerned in the initial phase rather than in the final, which means the public’s attention increases exponentially and results in a crazy spread at the beginning, nevertheless, there are few netizens participating in the discussion of the disposal result reported by the media.

For example, a male student in a southwestern university of China took upskirt photos of female students on 24th, August, 2019, being widely concerned in the network. The statistics of the Sina platform shows: the relevant information number of “College Students Sneakily Took Upskirt Photos of Girls in Library” in the entire Web is over 44,000 pieces, during 11pm, 24 ${}^{\text{th}}$ , Aug. to 3pm, 26 ${}^{\text{th}}$ , Aug. And by 3pm, 26 ${}^{\text{th}}$ , Aug., the related topics of it has exceeded 0.19 billion views and approximately achieved 5000 discussions.

From the below figure of information trend, it can be seen that the secret photo shoot event started to receive attention after being exposed to the net at 11pm, 24 ${}^{\text{th}}$ , August. On 25 ${}^{\text{th}}$ , the incident sprawled further after a news about the boy’s schoolmates protecting him in joint names, the remarks of school security, and the voice of the girl. The number of messages peaked at 5,968 pieces at 12 o’clock that day. Then the amount of messages began to drop with the small fluctuation caused by police’s report at 10pm.

Figure 2.

Information trend of a schoolboy in a southwestern university sneakily shot girls.

In addition, the influence caused by the network public opinion events usually spreads from a small scope to a large one and stretches from online to offline. What’s more, the influence of negative events is more serious than that of the positive events, also known as “Bad News Rule” [7]. Colleges and universities are primarily affected in two aspects: one is the influence caused by the improper handling toward the events; the other is the influence of not being responsible for the management and education in the events. While the influence on students mainly involves mentality and daily life. In comparison, when an individual student is at the heart of the online news, the student bears more heavier pressure in spirit and daily life.

Based on the college students’ reaction in the current public opinion events, psychology research, and research by some experts and scholars, it is concluded that the mental health of college students under the pressure of public opinion has the following characteristics:

Become frozen. The “frozen” state means that students present a blank in mind when faced with a tremendous pressure under a sudden condition [8]. Objectively speaking, the “frozen” state is a normal reaction against unexpected things, but in the environment of colleges, the longer the “frozen” state lasts, the more unfavorable to clarify facts and more harder to recover the mental and physical condition.

Be bigoted. Under the deliberate mass bombardment of tendentious opinions, students in the center of the events are easily losing objective and fair judgement and becoming obstinate. If the tendency is carrying immense pressure, the center person in the events will even take extreme reactions.

Turn to be closed in mind and body. As the influence expands gradually, the involved students will turn to be closed in mental and physical state. They feel overwhelmed, stressed out and lack courage to face the events. Then they keep themself in a closed space, escaping from the reality and society. Without a timely involvement of colleges, they are prone to have a heightened sense of isolation, alienation, and loneliness, which easily causes an abnormal personality.

Feel aggrieved, scared, and helpless. In the ferocious environment of cyberspace, emotional critique is always stronger than reasonable voice. A large proportion of netizens tend to focus on the public event itself instead of the thinking behind it. The fast transmission also adds difficulties to the related people to explain and clarify, for which negative moods such as grievance, fear, and helplessness are hardly avoidable.

What needs to be emphasized is that the specific event that causes the online incident is only a catalyst. For colleges, the fundamental methods of reducing negative consequences are to enhance precautions and management standards.

2. Mindsets and suggestions for school counselors to address the crisis of students

Instructors play a great role in the settlement of students’ crisis. Though they aren’t in the center of events, they are usually the key to control the development of the crisis. In general, there are three points for tutors to implement in order to solve the problems appropriately.

1)
Instructors need to investigate the causes at first time and report to the related departments in time, providing detailed information to help school make analysis and decisions. Although the Internet users’ personal emotions are more influential that the fact, the truth is still crucial for making decisions. The more concrete information they get, the more accurate and faster decisions can be made, which is highly valued in the following development of the events.
2)
The primary responsibility is to protect the mental and physical health of students. Tutors should assist students to get rid of the troubles with mental intervention [9]. In previous cases of psychological intervention, schools attached more importance on some specific and realistic catastrophic events while lacked research in mental intervention for both teachers and students. In fact, the potential network violence owing to the crisis is also likely to get the involved into distress, disturbance, even anxiety and desperation, or perhaps vegetative nerve symptom and behavior disorder [10]. Besides, it is quite necessary for counselors to protect students both mentally and physically and help them overcome the crisis on account that the online events usually happen abruptly, transmit fast, spread broadly, and some extreme net citizens even take extreme actions against the involved.

a)
Instruct students to have a good attitude. The spread of the net events has “News-hunting effect”, which means the cyber citizens are happy to satisfy their curiosity with the things proceeding rather than care about the truth. Therefore, in the beginning of the events, a “humble self-reflection” attitude is more easier than a clarification to ease the crisis.
b)
Measures should be taken to protect the safety of students. Colleges ought to take safety and stability as a priority. When students are stuck in the problems, schools has responsibilities to reduce the risk of safety of students and arrange someone to accompany them.
c)
It is important to pay attention to the mood change of the involved, supporting and enlightening them. Different from the normal mental intervention and considering the transmission rules of network, special notices are as follows:

i.
The uncertainty of the events’ development brings about obstacles in evaluating influential degree, making plans and obtaining feedback of mental intervention work [11].
ii.
Online citizens care more about the topics of the events. Under this condition, perhaps they ignore the specific situation that the involved are suffering. Thus, if necessary, preventing related students from connecting the cyber can stop continuous stimulation of negative information on the network.
iii.
In the practice of psychological intervention, instructors should stay neutral, try to understand, tell the involved they are acceptable and not comment on the event itself. The key is to ensure the safety of students [12].

3)
Counselor works as a bridge between the individual student and the remained classmates, to coordinate their relationship. Though the online events starts from a person, but the influence will stretch to surrounding students and schools. As a counselor, you need to make sure the safety of the involved, at the same time, also observe emotional reactions of rest students in the class, so as to avoid the atmosphere of isolation, anxiety and intensity 1̧3. When pacifying the class groups, instructors should explain the facts that are investigated and learned, from which students can know the events correctly and avoid distrusting each other. Moreover, Instructors need to highlight the specialty of individual case but not the universality of an entire group and guide students to form a right mentality to avoid similar cases.

3. Enlightenment of responding to the network public opinion crisis

When facing sudden events, colleges should take good advantage of “Primacy Effect” to occupy the high place of online public opinions that can influence the directions and trends of the events, clarifying truth to the public and seizing instruction right. That not only respects the public’s right to know, but also alleviates their curiosity, at the same time, maintains schools’ image and credibility.

In the paper, it compares the above mentioned college students’ online public opinion crisis with other references. Social online public opinion crisis handling plan is use for reference. Aiming at the peculiarities of college students’ mental health, it is concluded the research coping strategies based on innovative and comprehensive use of big data technology [14]. We establish a complete emergency protection system, which effectively protects the mental health of college students.

1) Apply network to form a positive impression against the negative influence of the events.

As the self media era comes, an increasing number of people use cell phones to express their opinions. Zhang Xiaolong, senior executive vice president of Tencent and president of Wechat business group, said in the Wechat open course PRO in 2021, Wechat users can have over 120 million new moments in every day. See the Fig. 3 to get more information.

Figure 3.

Data of Wechat moments in 2021.

The new media tool has become the first place of abrupt public opinions events. However, the events transmitted by the new media also exist problems like fragmented information, personalized emotions and so forth. Therefore, colleges can take good use of other network medias to quickly conduct positive promotions and reports, expressing their stands and values in some mainstream medias such as campus Sina Weibo, official Wechat account, school website, etc, so that the negative impacts can be reduced [15].

2) Try to change the trends of the public opinions events from the source.

The online public opinion events are of rapid spread. Netizens’ initial understanding of the truth only comes from the source of the events. These require a clarification from the source and a radical reverse of the trend [16]. Therefore, colleges need to analyze causes of the events, contact the involved person and units, and reach a consensus that aims to decrease the impact of the events by joint consultation.

3) Set up public opinions monitoring system.

The analysis system based on big data can receive warnings of the events timely and effectively, which generally includes five parts: data acquisition, data storage, data processing, data display, and data warning. These parts are respectively introduced as below:

Data acquisition usually applies the web crawler technology to regularly capture website data of Sina Weibo, Campus BBS, and Baidu post bar [17].

The web crawler can issue requests to websites and extract data needed from the response [18], generally by using the python library such as requests and beautifulsoup, etc.

Data storage refers to use distributed data library to store mass raw network data [19].

The distributed database can better store and manage a large amount of raw data, like MongoDB. This is the data storage way of the database:

Data processing is composed of data preprocessing and data analysis. The former mainly exerts the functions of extracting data, reducing noise and separating words, while the latter involves assembling the same category, analyzing emotions, etc [20].

In the preprocessing of data, it usually needs to extract features of context. The common algorithms include TF-IDF, term frequency-inverse document frequency, a common weighted technique used in information retrieval and text mining [21]. Its quantitative analysis formula as below:

$\displaystyle\textit{IDF}(x)=\log\frac{N}{N(x)}$ (1)

Among them, $N$ represents the total amount of texts in the corpus while $N(x)$ refers to the total amounts of texts including the word “ $x$ ” in the corpus.

The logistic regression is often used for text emotion analysis and public opinions events’ prediction [22], which is a regression model applied to narrow the range of prediction and limit the predicted value between 0 to 1 [23]. This algorithm introduces a sigmoin function on the basis of linear regression, as for the linear regression, the formula as follows:

$\displaystyle\theta_{0}+\theta_{1}x_{1}+,\ldots,+\theta_{n}x_{n}=\sum\theta_{i% }x_{i}=\theta^{T}x$ (2)

The predictive function in the logistic regression: values between [0, 1]:

$\displaystyle h_{\theta}(x)=g(\theta^{T}x)=\frac{1}{1+e^{-\theta^{T}x}}=\frac{% 1}{1+e^{-(\theta_{0}+\theta_{1}x_{1}+,\ldots,+\theta_{n}x_{n})}}$ (3)

Data display makes the data of the events visible, and the live display of the data can be showed through large screen and applications in smart phones.

The emergency system uses online public opinion crisis handling plans for reference, but the difference is that the collection of public opinion data is limited to the colleges. Figure 4 is an analytic system flowchart of massive big data public opinion. This process also borrows ideas from some online public opinion crisis processing systems.

The pre-warnings of public opinions events are instant notifications. The overall procedures of handling the data of the public opinions events operated via messages and notification of applications in cell phones are as following.

Figure 4.

System procedures of mass data analysis of public opinions events.

The above mentioned strategies based on big data analysis are briefly summarized as follows: The emergency system is established based on crawler data, big data storage and analysis, and big data visualization technology. The system will make intelligent judgments based on the preset threshold value [24]. When the public opinion data is above the critical value, the system will notify the relevant person in charge as soon as possible. And the relevant personnel will intervene at the first time and perform a second manual judgment. The system timely control public opinion.

4. Conclusion

With the development of information age, colleges’ tutors are facing more and more difficulties related to the online public opinions events in the management of students. The events are complicated and hard to handle. It requires not only instructors to have sharp insights, high disposal ability and the mastering of online transmission laws, but also colleges to establish emergency prediction, control the trend of the events and address the things properly. In addition, colleges also need to make instant reflections, conclusions, and researches, to solve the potential problems at the beginning, guide students to understand the crisis correctly, help them adjust mentality and physics, and jointly maintain harmony and stability of campus.

Footnotes

Acknowledgments

Fund project: Young Teacher’ Educational and Scientific Research Projects (Social Science) of Fujian Province in 2018; “Identification and Psychological Intervention for University Students who are Susceptible to Heresy” (Project Number: JAS180601), Psychological intervention team for college students in emergencies, Ningde Normal University professional team project for the academic staff, project number 2021XGTD08.

References

Chen

, eds. Public Opinion Study. Shanghai Jiaotong University Press; 2012.

Jia

Shi

. Strategies for universities against sudden public opinion events in campus. China-arab States Science and Technology Forum (Chinese and English). 2021 (08): 145.

. Analysis of impacts on university students’ value of public opinions events. University’s Communist Party Construction and Ideological Education. 2021 (14): 73.

. Guidance of ideological and political education of college students in the ecological perspective of online public opinions. The Guide of Science & Education(Mid-month). 2018 (05): 74.

Yuan

. Research on public opinion monitoring and guidance of college students in we media era. Journal of Hubei Open University. 2015; 28(12): 46. doi: 10.3969/j.issn.1671-5918.

Zhao

. A probe into the application of the principle of news communication in the events of network public opinions. Journal of News Research. 2020; 11(04): 69. doi: 10.3969/j.issn.1674-8883.2020.04.039.

Liu

. Transmission rules, guidance and control of emergent network public opinions. Shanxi Youth Journal. 2020; (17): 190.

Zhuang

Mei

Huang

. Psychological analysis of college students’ thoughts and behaviors affected by fake online public opinions under normal prevention and control. The Modern Occupation Education. 2021; (06): 1. doi: 10.3969/j.issn.1672-8637.2019.08.027.

Feng

. Guidance methods selection of network public opinion of counselors from the perspective of media convergence. Times Report. 2021; (06): 159.

10.

Zhao

. University Counselors’ Role Positioning and Ability Improvement in the Instruction of Online Public Opinion Events: A Case Study of Public Health Emergencies. Journal of Multimedia and Network Teaching in China, (in the first ten days of a month). 2021; (04): 120.

11.

Zhao

. Viewing enterprise public opinion events from the perspective of psychological crisis intervention. The Economy. 2019; (08): 97.

12.

Wang

. The Role Positioning and Research on Competence of Counselors in Universities’ Psychological Crisis Intervention. Cloud Forum for Course Teaching and Management in 2020 (Gui Yang), Volume of Papers: 74.

13.

. Strategic research on guidance of universities sudden public opinions events in the new era. Writers World. 2020; (07): 88.

14.

Zhu

Yao

Wang

. The influence of Live Stream (We Media) on College Students. Chinese & Foreign Entrepreneurs. 2018; (23): 168.

15.

. Some thoughts on universities’ response to the emergent network public opinion events in the era of new media. Policy Research & Exploration (Chinese). 2021; (07): 86.

16.

. Research on practical path of university network ideology in the new era. Journal of Changchun Normal University. 2021; 40(07): 125.

17.

Wang

Zhao

. Design and implementation of college network public opinions analysis system based on crawler. China Computer & Communication (Theoretical Version). 2021; 33(03): 137.

18.

Zhang

. University Network Public Opinion Early Warning Research based on LSTM-CNN Model. Journal of Yulin University. 2021; 31(04): 44.

19.

Huang

. The application of big data in information management system. Computer Programming Skills & Maintenance. 2021; (05): 100. doi: 10.3969/j.issn.1006-4052.2021.05.041.

20.

. Early warning research on theme alienation of emergencies in the environment of public opinion big data. Network Security Technology & Application. 2021; (08): 32. doi: 10.3969/j.issn.1009-6833.2021.08.020.

21.

Cao

Sheng

Zhou

. Research on Extracting News Keywords based on TF-IDF-MP Algorithm. Journal of East China Jiaotong University. 2021; 38(01): 122.

22.

Yang

. Research on data storage and data security. Network Security Technology & Application. 2021; (03): 54. doi: 10.3969/j.issn.1009-6833.2021.03.031.

23.

Feng

Wang

. Analysis and research of micro-blog emotion based on logistic regression model. Computer & Digital Engineering. 2018; 46(09): 124. doi: 10.3969/j.issn.1672-9722.2018.09.026.

24.

Zhou

. Management and guidance of university cyber public opinion events under network power strategy. Network Security Technology & Application. 2021; (07): 115. doi: 10.3969/j.issn.1009-6833.2021.07.065.