China in the presence of Weibo
Literature cards
# Editor's note
At present, public discussion in cyberspace presents the characteristics of interweaving entertainment and sociality, and the theme shift and long-tail effect of issues in the field of public opinion are prominent. There are some problems in the related discussions in the field of communication, such as single-orientation, logical fracture, and lack of in-depth consideration. Dong Chenyu and Lin Qihuan tried to transfer the analytical thinking of "theme evolution" to communication studies in the 6th issue of Media Observation in 2024, selected the representative event of Internet celebrity Li Jiaqi's "angry netizens", conducted social network analysis from the three levels of theme, emotion and behavior, and discussed the characteristics, correlation and evolutionary logic of each node under the advancement of time context across stages. It is found that the theme, emotion and behavior show a linkage deviation trend, showing the characteristics of horizontal jump and emotional fluctuation of behavior, but there is also a continuity of network structure in polarization and tearing. In the process of evolution, emotions and behaviors play the roles of "adhesive" and "metronome" respectively, and the evolutionary structure of the theme presents a composite state of interweaving time and space. Inspired by the above research process, this paper proposes a new idea of network analysis based on integrated and co-evolution, that is, an analytical perspective of horizontal and vertical integration, inter-network collaboration, and dynamic evolution.
# Essence of Ideas
1
The concept of thematic evolution has a natural connection with journalism and communication, which was born to track news reports.
2
In the evolutionary structure of the theme, emotion plays the role of "glue", gluing the various elements scattered in the space of public opinion.
3
In the evolutionary structure of the theme, behavior plays the role of a "metronome", regulating the rhythm of the evolution of the theme and emotion.
4
The interaction between technology carriers and social emotions may make the thematic evolution of Internet events in the Chinese Chinese context more complex and meaningful.
# Keywords
thematic evolution; social network analysis; Internet celebrity industry; LI Jiaqi; Emotion; behavior
# Citation format
(1) Dong Chenyu, Lin Qihuan. "Adhesion" and "Metronome": The Theme Evolution of Online Public Opinion Issues: A Multi-dimensional Social Network Analysis of Li Jiaqi's "Angry Netizen" Incident[J].Media Observation,2024(06):24-34.)
(2) Dong Chenyu, Lin Qihuan. (2024).“ "Adhesion" and "Metronome": The Theme Evolution of Internet Public Opinion Issues: A Multidimensional Social Network Analysis of Li Jiaqi's "Angry Netizens" Incident. Media Watch (06), 24-34.
The body of the paper
"Where's the price?" Li Jiaqi in the live broadcast frowned, responding to netizens' comments that the price of a certain brand of eyebrow pencils at 79 yuan was too high, "Sometimes find your own reasons, have your wages risen after so many years, have you worked hard?" Soon after, the topic of "Li Jiaqi brought goods and angered netizens" rushed to the hot search on Weibo. According to the "2023 Weibo Hot Trends Report", "Li Jiaqi's inappropriate speech controversy" ranked second in the celebrity celebrity hot social event index for the whole year, second only to "Coco Li's death".
It is worth noting that in the Li Jiaqi incident, the "matter-of-fact" discussion is only the tip of the iceberg. The generation of issues is like the process of tornadoes forming "climate", which is the result of multiple factors checking and balancing each other. With the passage of time and the involvement of more subjects, the center of the topic is constantly changing: the focus of netizens has shifted from the Li Jiaqi incident itself to derivative issues such as employment, income, domestic products, and live broadcasts, among which there are many emotional and clear views. What's more, a similar "long-tail effect" is prevalent in today's online public events. So, how to structurally understand this phenomenon of topic migration? How do you understand the motivation and logic behind migration?
In order to answer this question, this study uses the social network analysis method to sort out the phenomenon of topic evolution by taking the Li Jiaqi incident as an example. In a nutshell, this paper conducts horizontal and vertical network construction, central computing and QAP analysis in three stages, aiming to summarize the concrete representations of the long tail of the event, discuss its motivation and logical relationship, and propose a new perspective of network analysis based on this.
1. Literature review
(1) Theme Evolution: An Interdisciplinary Concept and Method Application
"Theme evolution" refers to the phenomenon of gradual change of topics in the process of communication, including the replacement of old and new topics in time and the diffusion/convergence of themes in space. Referring to the concept of "evolution" in biology, theme evolution has "heritable variation", which is mainly divided into three modes: content evolution, intensity evolution and structure evolution. The concept of thematic evolution has a natural connection with journalism and communication, which was born to track news reports. However, at present, this concept is mainly applied in the fields of information management and computer science, and related research focuses on the exploration and updating of thematic evolutionary analysis tools and techniques, while communication scholars rarely use this concept directly.
Even so, this does not mean that communication studies are indifferent to the flow of information in the process. Some communication scholars start from the "transmission mechanism" and take the flow of rumors and rumors in the process of transmission as the research object. For example, Liao Mengxia described the development of rumors during the epidemic as a "snowball" rolling, presenting a multi-thematic situation under the construction of language families of different sizes and contents. Other communication scholars refer to the phenomenon of theme evolution as "public opinion evolution" and "public opinion generation and diffusion", and have investigated the different stages of public opinion development. For example, Li Fei et al. believe that the communication of new media public opinion has gone through an embryonic period, a gestation period, a fermentation period, and an explosive period from linearity to fission.
In a word, the current research on online public opinion is biased towards the analysis of the technical level, showing an obvious tendency of technocracy. This also leads to the research on "theme evolution" from the perspective of information management and computer science with strong instrumental attributes, focusing on the methodological path of network public opinion monitoring and response. In contrast, the field of journalism and communication rarely uses this concept directly, and the related discussions are characterized by diverse concepts and vague expressions. Therefore, this study attempts to combine interdisciplinary perspectives, transfer the analytical orientation and method of "topic evolution" to communication studies, and construct a framework that is more suitable for network analysis of social communication phenomena.
(2) Emotional Evolution and Behavioral Evolution: The Concept of Theme Linkage
In order to achieve this, we need to first overcome the mechanical nature of the field of information science and computer science in dealing with the topic of "topic evolution". Specifically, this article adds emotional and behavioral attributes in addition to the informational attributes of the topic. These two parts are still largely absent from the existing thematic analysis.
On the one hand, the change of public opinion does not only depend on the facts of the event itself, but is also driven and influenced by external emotional factors. In the context of the Internet, digital media has changed the state of emotional circulation, and emotions accumulate in communication. Emotion is a level that cannot be ignored when focusing on the evolution of the theme. After the formation of emotion, it enters the stage of diffusion and flow, which seems to be chaotic and disorderly, but in fact it is a cultural process affected by social norms and other factors. At present, there have been limited discussions on the process of emotional evolution that accompanies the evolution of themes, but this part of the research generally deals with emotions in a more mechanical stage and model. Similarly, many studies have used emotion as a static mediator to connect information and behavior, and have not addressed the dynamic evolution of emotion.
On the other hand, in addition to sentiment analysis, the analysis of users' behavior is also essential to understand communication in social media. At present, the analysis of netizens' behavior mainly focuses on their "interactive" behavior, such as retweeting and commenting on Weibo, initiating and responding to topics, etc. These studies often associate netizens' communication behavior with communication intention, or regard the accumulation of comments and forwarding as the driving force for the formation of public opinion. In contrast, the existing research focuses more on the appearance of user behavior, and lacks attention to the internal motivations, attitudes and positions of user behavior.
It can be seen that most of the current discussions related to topic evolution have four problems: first, most of the studies only focus on the intensity evolution of "theme evolution", and mainly focus on the more mechanical stage presentation, rather than the specific exploration of its structural change process; Second, although some scholars have tried to make a specific analysis of the evolution of information structure, the field they focus on is still limited to the relatively closed WeChat group, and does not involve the more complex and changeable field of public opinion. Third, the existing research on emotions and behaviors is relatively weak, such as simply classifying emotions as positive or negative, or focusing only on the user's behavioral representations and ignoring their deep intentions, and lacking in-depth consideration of details. Fourth, theme, emotion, and behavior are in fact a unified whole, which may trigger and transform each other, but most of the existing studies discuss these three separately and lack the efforts to connect them.
Based on the above discussions, this paper attempts to transfer the analytical thinking and methods of "topic evolution" to communication research. In terms of methodology, based on the horizontal and vertical social network analysis of the evolution of specific public opinion events, on the basis of spreading out the theme, emotion and behavior network, combined with quantitative data analysis and qualitative text analysis, the characteristics, correlation and evolution logic of each node under the passage of time are discussed in a cross-stage and hierarchical manner, and a network analysis idea of integrated and collaborative evolution is tried to be developed accordingly (see Table 1). In terms of research objects, we identified the Internet celebrity industry at the intersection of entertainment and society as the entry point, and selected the Li Jiaqi incident for analysis.
(3) Why Li Jiaqi: Internet celebrity symbols on multi-dimensional intertwined points
"Influencer" is a development of the traditional concept of "celebrity". Different from the concepts of "micro celebrities" and "online influencers" commonly used in Western academic circles, "Internet celebrity" is a Chinese localization expression with the special localization characteristics of the Chinese model. From a historical perspective, since Taobao first proposed the concept of "Internet celebrity economy" in 2015, the frequency of use of the word "Internet celebrity" in public discourse has shown a blowout growth, and the popularity of discussion even surpassed that of traditional celebrities for a time. Nowadays, the boundaries between Internet celebrities and celebrities in terms of individual influence, profit methods, and performance models are becoming increasingly blurred, showing a trend of "Internet celebrity stardom". At the same time, "Internet celebrities" are closely related to society, culture and capital, and have a significant influence in public discourse and business discourse, and are an emerging group at the intersection of society, economy and culture.
Specifically, the reason why this study chose the Internet celebrity-related incident of "Li Jiaqi angrily scolding netizens" for case analysis is that, on the one hand, as a Chinese phenomenon-level Internet celebrity, Li Jiaqi has become a symbol, representing an identity transition from grassroots to Internet celebrity and then to star, condensing people's complex imagination of Internet celebrities; On the other hand, as a social action node, Li Jiaqi connects the public, businesses, platforms and other subjects in society, and his role and identity imply multiple contradictions and conflicts. After the incident of "Li Jiaqi angrily scolded netizens", it quickly became the focus of discussion, and continued to maintain a high degree of popularity in the follow-up fermentation.
In addition, in terms of public opinion characteristics, this incident has an obvious "long-tail effect". In other words, the focus of public opinion on this incident has shown an obvious theme shift and divergence: the focus of netizens' discussion has spread from the incident itself to the discussion of domestic products and the live streaming industry. It also provides a suitable sample for observing the evolution of themes, emotions and behaviors. Therefore, the incident is not only limited to the industry problems of the Internet celebrity industry, but also implicates a series of social issues; Its communication process contains the fermentation logic of public opinion on entertainment events under the trend of Internet celebrity stardom. The analysis of the trend of public opinion in this event can directly present the bridge and collusion between the entertainment and social nature of public opinion. Specifically, this paper will focus on the following research questions: 1. What is the thematic evolution process of Li Jiaqi's angry netizen incident? In the process of the evolution of this theme, will the emotions and behaviors of netizens also evolve? 2. What are the specific representations of the evolution of the theme, emotion, and behavior of the event? 3. From Internet celebrity Li Jiaqi to a wider range of public opinion, what is the motivation and logic of the evolution of the theme?
2. Data collection and research methods
(1) Data collection
1. Study sample selection
As one of the largest social media platforms in China, Weibo is the core area of social participation and discussion in China's cyberspace, which not only accommodates grand or specific discussions among various social subjects, but also records the evolution of social hot events. As a node in cyberspace, Weibo hot search entries are a direct embodiment of the evolution and diffusion of hot events. Therefore, this study selects Sina Weibo as the research carrier, and takes the 2023 "Li Jiaqi Live Broadcast Angry Netizens" and a series of derivative hot search topics as the research object, and conducts a specific analysis of the hot search entries on Sina Weibo and the narrative texts of social media opinion leaders and the public under the entries.
Specifically, the main data for this study comes from a joint research project between Weibo and the Institute of Journalism and Social Development of the Chinese University in January 2024. According to the development nodes of the data, we preliminarily divided the event into three stages, which were used as the sample extraction time interval for this study. The first stage is related to the information and details of the "Li Jiaqi live broadcast with goods and angry netizens" incident, and the time period is from September 10 to September 12, 2023; The second stage is a broader discussion around the incident of "Li Jiaqi's live broadcast of goods and angry netizens", involving domestic products, social issues, etc., with a time period from September 13 to September 16, 2023; The third phase mainly focuses on issues such as the dispute over the monopoly of low-price live streaming, and the time period is from October 24 to October 26, 2023.
2. Data collection and cleaning
On Weibo, each Weibo post is a relatively independent unit of meaning, and there may be problems such as duplication of content and lack of substantive meaning between posts. As a collection of posts, the topic includes different posts and comments around the topic, which has a large amount of data and is difficult to clean. Therefore, this paper mainly uses Weibo topics as the data collection unit, but also refers to the posts and comments under the topic to summarize the topic and other information when coding. After collection, the researchers preliminarily obtained 139 hot search words on Weibo.
In the process of further combing and cleaning the data, a total of 7 entries in the following three situations were eliminated: one was that the topic involved violations were not displayed (such as "the arrogance of the new elite", etc.); the second is a long-term fixed entry that involves the subject of the event, and contains a large amount of information that is not related to the event of concern in this study; The third is unnatural fermentation, which is obviously a design entry with marketing purposes (e.g., "Jingdong 8 p.m. live broadcast to buy goods and talk to the boss"). After the data cleansing was completed, the official total sample size was 132. Among them, there were 60 samples in the first stage, 42 samples in the second stage, and 30 samples in the third stage.
(2) Data coding and analysis
1. Develop a coding table
In order to observe the trend of theme deviation, emotion and reaction in the incident of Li Jiaqi's angry netizens, this study will encode the theme, emotion and behavior of the collected topics, and set several attributes under each level. Due to the strong particularity of this topic, and most of the content involved is strongly related to Li Jiaqi and specific domestic products, the researchers have repeatedly sorted out and discussed the obtained samples, and summarized 10 theme attributes: social issues, historical issues, live streaming, price, domestic products, Li Jiaqi and related parties, other anchors/live broadcast rooms, other social personnel, responses, and popularity (see Table 2). At the emotional level, the researchers referred to the two emotional polarities and eight basic emotions in the NRC Emotion Lexicon proposed by the National Research Council of Canada, as well as the seven types of emotions summarized in the Chinese emotional vocabulary ontology database of Dalian University of Technology, and sorted out six emotional attributes: anger, sadness, evil, surprise, goodness, and happiness (see Table 3). At the behavioral level, due to the desire to investigate more specific behaviors and tendencies, this study does not refer to the existing research to simply divide behaviors into forwarding, commenting, etc., but integrates the behavior performance of netizens in the posts and comment areas under the topic, and summarizes five behavioral attributes: support, opposition, questioning, rational discussion, and entertaining discussion (see Table 4).
2. Coding
In this study, 132 samples of Weibo hot search entries were encoded, and the coding work was completed by two coders, and a consensus was reached on the specific meaning of each coding category before coding. Due to the extremely mixed and complex views in the Weibo topic discussion field, if a variety of themes, emotions, and behaviors are presented in the topic when coding, they will be recorded at the same time, such as the topic that the cost price of the same eyebrow pencil of the same brand involved in Li Jiaqi's live broadcast does not exceed 10 yuan, which involves both domestic products and prices. At the same time, 10% of the total sample size (N=13) was taken by the two coders for reliability testing, and the reliability calculation of the two coders at the three levels of subject, emotion and behavior was more than 0.80 according to the Kribendorff α coefficient formula, and the reliability α of the two coders at the three levels of subject, emotion and behavior were all greater than 0.80, which met the requirements. The two coders then distribute the samples evenly and complete the subsequent coding work.
3. Data analysis methods
Social Network Analysis uses visualization technology to visualize social networks in an intuitive form, making them easy to observe and helping researchers discover their characteristics and patterns. After the coding, the researchers used the social network analysis tool UXINET to conduct mapping and data analysis to depict the theme, emotion and behavioral characteristics of "Li Jiaqi's live broadcast angry netizens" at different stages of topic development.
Specifically, the researchers first counted the number of co-occurrences of the subject-emotion and subject-behavior attributes in each stage, made a two-mode table to draw a network relationship structure diagram, and calculated the degree centrality of each attribute to determine which topics were the focus of discussion and which emotions and behaviors dominated in a particular stage. Secondly, this study converts the above-mentioned two-mode table into a one-mode table, and uses the quadratic assignment procedure (QAP) to verify whether the theme, emotion, and behavior structures in the three stages are related, so as to observe the continuation of the network structure of the three levels over time. In order to observe the changes and correspondence between emotions and behaviors more intuitively, the researchers further extracted emotions and behaviors separately, counted the co-occurrence of topics under different emotions and behavior attributes, and also made a network relationship structure diagram and conducted QAP analysis in stages (see Figure 1).
3. Research findings
(1) Theme, emotion, and behavior linkage shift
In order to answer the research question 1, this study first verified the theme shift in the Li Jiaqi incident through data analysis and made it intuitive. Furthermore, the researchers observed the theme, emotion and action in tandem through the network structure diagram, and summarized the linkage of the three in the process of excursion.
The researchers counted the co-occurrence between theme and emotion, theme and behavior in each stage, and calculated their degree centrality (see Table 5 and Table 6). This concept reflects the centrality of nodes in the network. The study found that at each stage, the top three topics pointed to by emotions and behaviors, i.e., the focus topics, were the same: the first stage was "Li Jiaqi and related parties" (degree=9.00), "domestic products" (degree=9.00), and "price" (degree=6.50), and the second stage was "domestic products" (degree=8.83), "price" (degree=4.17), and "live streaming" (degree=1.83). The third stage is "live streaming" (degree=6.33), "other anchors/live broadcast rooms" (degree=5.33), and "Li Jiaqi and related parties" (degree=5.17). Therefore, we can intuitively observe the transformation of the central theme in the three stages, that is, it proves that there is a thematic shift in the discussion around the Li Jiaqi incident in the Weibo public opinion field: the focus has shifted from Li Jiaqi and his related parties to domestic products, and then to the issue of live streaming.
The researchers further carried out network visualization mapping of the co-occurrence of subject-emotion and subject-behavior attributes (see Fig. 2 and Fig. 3), so as to intuitively observe the correlation and interaction between theme, emotion and behavior at each stage.
The study found that in the first stage, the main emotions of netizens were evil and anger, and their behaviors were mainly opposed, and they all had the strongest correlation with Li Jiaqi and his related parties, such as "why not eat minced meat, respect and a sense of boundaries", "Do you accept Li Jiaqi's apology" and other topics appeared in the discussion. In the second stage, the main focus of contradictions shifted from Li Jiaqi, the spearhead of public opinion was mainly directed at domestic products, emotions and behaviors showed contradictions and mixed situations, evil, good, happiness, and anger all occupied the center, and behavior also showed a situation of coexistence of support and opposition; Specific topics include "what are the domestic brands to be proud of" guided by positive emotions, and "how many Western dollars are left after your salary is reduced by 3,000" led by negative emotions. In the third stage, the focus of public opinion shifted to live broadcasting, and in the process of shifting, more subjects were involved except Li Jiaqi and domestic products, and the central emotions returned to negative emotions such as evil and anger, while the positive emotions of good and happiness slipped to the end, and the main behaviors were opposition and rational discussion.
(2) Figurative representation: ups and downs, polarization and continuation
In order to answer research question 2, this study summarizes the concrete representations of themes, emotions and behaviors in different stages, centrality calculation, and QAP analysis in the evolution process. The study found that in the process of evolution, there were not only horizontal jumps in behavior and emotional fluctuations, but also in a specific stage, which would also be torn into polarized public opinion tendencies. However, despite the continuous diffusion, evolution and shifting, the thematic axis still exists and is overlooked, making the evolution of themes and emotions, themes and behaviors continuous.
The first is behavioral jumps and emotional ups and downs. Calculating the centrality of the behavior in the three stages of subject-behavior co-occurrence, it can be seen that there is a "horizontal jump" of behavior in the cross-stage process: the behavior in the center of the first stage is "opposed" (degree=10.60), but the behavior in the center of the second stage jumps horizontally to "support" (degree=3.70), and the behavior "opposed" (degree=5.70) in the third stage becomes the dominant behavior again.
In addition, it can be seen from the ranking of the centrality of each emotion in the theme-emotion co-occurrence (Table 5) that in the first two stages, the emotional "good" and "happy" ranked second and third, respectively, occupying a more central position. Analyzing the co-occurrence of the themes corresponding to these two positive emotions, it can be seen that the theme with the highest central value in the first two stages is "domestic products", that is, the discussion of these two emotions mainly revolves around "domestic products". When the centrality of "rational discussion" in the behavior rises from the middle and late stages of the first two stages to the second place in the third stage, and the core of the theme shifts from "domestic products" to "live streaming", the centrality ranking of "good" and "happy" also slides to the end. It can be seen that when rational discussion returns and the topic shifts to a new direction, emotions also fluctuate. It is important to note that there are no emotional or behavioral attributes that are always on the periphery or at the center in the different stages. Therefore, as time progresses, when the theme shifts in different directions, the emotions and behaviors in the field of public opinion are also constantly rolling and changing.
The second is emotional polarization and tendency to tear apart. By analyzing the co-occurrence of themes under emotional attributes, this study found that in the second stage, there was a positive correlation between "good" and "mourning" (r=0.49, P<0.01) and "happiness" and "mourning" (r=0.57, P<0.01). In the third stage, there was a positive correlation between "good" and "evil" (r=0.39, p<0.05). In other words, at a given stage, the structure of the thematic network between these emotions is similar, and the topics to which they point are highly consistent.
However, these three sets of emotions are contradictory. For example, "evil" with obvious negative tendencies and disgust and "good" with obvious positive tendencies and affection and trust appear at the same time in the same topic discussion field. Specifically, in the third stage of the discussion on the topic of "Hai's Electric's Response to Real-Name Report on JD.com", Hai's Electric reported that JD.com was suspected of low-price monopoly. This incident directly involves both sides,Netizens often stand in line directly in the discussion,Some netizens support JD.com,Think this is "the platform sticks its own money",But others support Hai's Electric,"This is the skin of the manufacturer",As a result,There are two kinds of torn contradictory emotions in the coding results"Evil"And "Good"At the same time. It can be seen that compared with the more unified emotional tendency in the first stage, when the topic shifts and involves more diverse subjects, there may be more obvious emotional polarization and tendency tearing in the public opinion field.
The third is similarity and continuity. It is worth noting that when we calculate the centrality ranking of each emotion in the three stages of theme-affective co-occurrence, we find that the emotion that ranks first in each stage is evil or angry. It can be seen that in the Li Jiaqi incident, the central emotion throughout is negative emotion.
The results showed that the topics in the first stage were significantly positively correlated with the topics in the second (r=0.51, p<0.05) and third (r=0.54, p<0.05) stages, respectively. It can be seen that although there is a shift in the theme in the process of the three stages, there are still similarities between the theme network structure of the first stage and the second and third stages, and the theme content has continuity.
In addition, the researchers also extracted different attributes of emotion and behavior separately, counted the co-occurrence relationship of topics under different emotion and behavior attributes, and calculated the QAP correlation relationship of different emotion and behavior topic networks at different stages. After calculation and collation, it is found that there is a significant positive correlation between "rational discussion" and "questioning" in the three stages. It can be seen that the topic structure of rational discussion and questioning is similar in the three stages, which shows that netizens have always maintained a certain rational attitude when questioning the topic.
IV. Discussion and Conclusions
This paper attempts to transfer the analytical thinking of "theme evolution" to communication studies, selects the highly representative event of Internet celebrity Li Jiaqi's "angry netizens", analyzes the social network from the three levels of theme, emotion and behavior, and discusses the characteristics, correlation and evolution logic of each node under the advancement of time context across stages. It is found that the theme, emotion and behavior show a linkage deviation trend, showing the characteristics of behavior horizontal jump and emotional fluctuation, but there is also a continuity of network structure in polarization and tearing. In the discussion part, the study will try to shift from the case of Internet celebrity Li Jiaqi to a broader field of public opinion, discussing the motivation and logic of the evolution of the theme. In addition, the methodological potential of the article is also described.
(1) "Adhesion" and "Metronome": Emotional and behavioral roles in the evolution of themes
In the evolutionary structure of the theme, emotion plays the role of "glue", gluing the various elements scattered in the space of public opinion. At present, the "sense of vulnerability" is spreading in China's public space, and this feeling has expanded from the meaning of "loss of employment" and "hardship in life" to the meaning of social differentiation such as "sense of deprivation" and "insufficient protection of rights". When Li Jiaqi asked "where is it expensive" in the live broadcast room, standing on the opposite side of the social emotional structure, this long-term accumulation of powerlessness and injustice and the general "sense of vulnerability" was instantly ignited. Netizens questioned "what does 79 yuan mean to ordinary people", grafting anger and disgust onto the Li Jiaqi incident, triggering the first stage of discussion about Li Jiaqi. In the long tail of the topics in the next two stages, whether it is the discussion around domestic products and prices, or the turmoil of the low-price monopoly of live streaming goods, the general emotional structure of society has set the tone for the theme of public opinion, and "anger" and "evil" have always run through the discussion of various topics, and have become the driving force for behavior to appear and the theme to continue to roll forward. It can be seen that emotion as an "adhesive" glues together fragmented themes or behavioral elements in horizontal space; This state of adhesion also forms a linkage force, which leads the evolution and change of the theme and behavioral structure on the vertical timeline.
In the evolutionary structure of the theme, behavior plays the role of a "metronome", regulating the rhythm of the evolution of the theme and emotion. The performance in the network is often an incomplete "mapping" of the real society, and the contradictions of all parties mixed with emotional factors are expressed in the online public opinion, which will deviate from the actual situation of the real world to a certain extent. Among the behavioral attributes that this study focuses on, rational discussion and questioning often occur at the same time, and netizens think and analyze around specific problems without easily falling into the emotional whirlpool, which slows down the evolution process. For example, in the third stage, when netizens made a rational analysis of issues such as the "floor price agreement", the public opinion field showed a cooling state. The behaviors of support or opposition are always at the core of netizens' behaviors, and they have obvious attitude biases and show emotional representations, which trigger more controversy and interaction, which is the accelerator of the evolution process. For example, in the discussion of whether to support domestic products and rub the heat of the Li Jiaqi incident, netizens quickly stood in line, and there was no shortage of sabre-rattling views between yes or no, and the whirlpool of public opinion expanded rapidly. Entertainment discussions, such as the production of Li Jiaqi's "Which Li is expensive" emojis, etc., are detached from serious and rational discussion and analysis, and quickly promote the evolution of public opinion themes and emotions, which is also more likely to cause deviation and distortion results. As a result, under the control of tension and relaxation, which is the "metronome" of public opinion, the deflection velocity of the structural network is also adjusted.
(2) The composite evolutionary structure of spatio-temporal interweaving
Thematic evolution has the characteristics of "heritable variation", which has both the stability of "heredity" and the difference of "variation". Evolution is not only a simple staged theme transfer or emotional and behavioral transformation at the information level; It is not only the snowballing range diffusion and intensity accumulation at the intensity level. The process of "evolution" constitutes a criss-crossing multiple network structure, which can be dismantled from two dimensions: space and time.
Spatially, the elements of public opinion present a state of continuity in which the center is prominent. As the network structure spreads out, the elements are connected to each other, and there is a relationship of adhesion and inheritance. First of all, the shift and evolution of the theme is not an atomization process that diverges at will, and the theme is not randomly "scattered", nor is it irregular, but always presents a central network. For example, in the incident of Li Jiaqi's angry attack on netizens, from "Li Jiaqi and his related parties", to "domestic products", and then to "live broadcast with goods", the topics that occupy the core position in the three stages have bridged nodes with the initial incident, and new topic branches have been extended along with the angry or happy emotions in cyberspace. Further dismantling of the elements of theme, emotion or behavior shows that the attribute categories throughout several stages are basically stable, that is, the fulcrum in the network structure is always consistent, but the relative relationship and subordinate position between each fulcrum have shifted. In the network diagram of the three stages of the Li Jiaqi incident, although the centrality of attributes such as "domestic products", "prices" and "social issues" fluctuate with the progress of the stages, they always exist in the public opinion space as the fulcrum of the leveraged issues.
In time, the evolution of themes, emotions, and behaviors appears to be non-linear, easily triggered. The evolution process does not progress at a uniform pace with the linear timeline, but presents a leap and continuity under the connection of emotions and the rhythmic adjustment of behavior. A topic may disappear for a period of time, encounter a new "public opinion thread", be dragged to bridge and combine with another event, and thus be detonated again. For example, in the course of the Li Jiaqi incident, the discussion about Li Jiaqi and domestic products gradually fell silent in mid-to-late September, but when Jingdong purchased and sold "shouting" Li Jiaqi at the end of October, exposing his suspected price monopoly, it quickly set off a new round of public opinion for a while, becoming a continuation of Li Jiaqi's angry netizen incident, Internet celebrities Brother Yang, Brother Xiaoyang, Hai's Electric, Li Jiaqi's company Meione and other subjects were gradually involved, and netizens discussed the low-price monopoly of live broadcasts. Therefore, the evolution of the theme does not present a unified and uniform state at the temporal level, but under the staggered influence of emotions, behaviors, and new public opinion fulcrums, it is staggered and fast, jumping and falling in the continuation.
From a more macro perspective, China's current public opinion ecology has greatly exacerbated this staggering and undulating, making the long tail of online public events longer and more unpredictable. On the one hand, Chinese society is in a critical period of reform and transformation, which will inevitably be accompanied by constantly updated social contradictions. On the other hand, the public's sense of subjectivity and demand for expression are also becoming increasingly strong. In this state, the Internet undoubtedly carries the emotional fragments of different groups in society. These fragments use online public events as carriers, embedding them into them, and then "assembling" different themes. For example, in 2023, the incident of "Coco Li's death" sparked people's discussion of depression, which in turn triggered online public opinion condemnation of a variety show; The incident of "Ningxia farmers kneeling on the ground begging for water" has sparked a voice for social justice, as well as discussions about desertification and planting trees to control desertification. In short, the interaction between technology carriers and social emotions may make the thematic evolution of Internet events in the Chinese Chinese context more complex and meaningful.
(3) Network analysis of integrated co-evolution: a method exploration
The above analysis of the evolution of the public opinion theme, emotion and behavior of the Li Jiaqi incident, as well as the summary of the interaction law of the three in the public opinion field of Weibo, is an application attempt from an interdisciplinary perspective, which verifies the possibility of transferring "theme evolution" to communication studies: the dismantling, analysis and interpretation of the public opinion field from the perspective of horizontal and vertical combination can observe the continuity of the theme in the shift, and summarize the interactive logic of the theme, emotion and behavior in the evolution process; Starting from different thinking axes, the observed spatiotemporal characteristics of the network are also different. This kind of transfer application from an interdisciplinary perspective helps us to cross the rupture of "theme-emotion-behavior" in the field of public opinion. Therefore, this paper hopes to further extract and condense a new perspective of network analysis based on integrated co-evolution.
The concept of co-evolution comes from the idea of biological mutual benefit, that is, populations can interdependent, interact, and adapt, rather than compete with each other or atomize each other. This perspective focuses not only on the evolution of interactions between networks, but also on the interactions between individual elements and the networks to which they belong. Inspired by this concept and verified by the above empirical process, this paper proposes the perspective of integrated co-evolution network analysis. This perspective is horizontally and vertically integrated and linked, using quantitative methods to summarize the rules and qualitative ideas to explain the causes, so as to pay attention to the different aspects and levels of the network, and avoid the problems of instrumental mechanical rationality or pure speculation and lack of foundation. This perspective is synergistic, rather than atomized and isolated, and not only focuses on the interaction between elements and between elements and networks, but also observes the interaction between the whole network and another network. This perspective jumps out of static spatialization thinking and phased division, and focuses on the specific network evolution dynamics and interweaving processes in temporality. At a time when global relations are complex and intertwined, and connections and divisions coexist, when the wave of modernity brings more "uncertainties" to be asked, it is hoped that this integrated, collaborative and dynamic research perspective can bring some new possibilities for the new development of journalism and communication disciplines.
(Published in the "Presence" column of Media Observer, Issue 6, 2024, notes omitted.) Please refer to the original text for academic citations. ）