Chinese Microblog Research Articles

A comparative analysis for spatio-temporal spreading patterns of emergency news

Understanding the propagation characteristics of online emergency news communication is of great importance to guiding emergency management and supporting the dissemination of vital information. However, existing methods are limited to the analysis of the dissemination of online information pertaining to a specific disaster event. To study the quantification of the general spreading patterns and unique dynamic evolution of emergency-related information, we build a systematic, comprehensive evaluation framework and apply it to 81 million reposts from Sina Weibo, Chinese largest online microblogging platform, and perform a comparative analysis with four other types of online information (political, social, techs, and entertainment news). We find that the spreading of emergency news generally exhibits a shorter life cycle, a shorter active period, and fewer fluctuations in the aftermath of the peak than other types of news, while propagation is limited to a few steps from the source. Furthermore, compared with other types of news, fewer users tend to repost the same piece of news multiple times, while user influence (which depends on the number of fans) has the least impact on the number of reposts for news of emergencies. These comparative results provide insights that will be useful in the context of disaster relief, emergency management, and other communication path prediction applications.

An Experimental Study of Spammer Detection on Chinese Microblogs

With the development of Web 2.0, social media such as Twitter and Sina Weibo have become an essential platform for disseminating hot events. Simultaneously, due to the free policy of microblogging services, users can post user-generated content freely on microblogging platforms. Accordingly, more and more hot events on microblogging platforms have been labeled as spammers. Spammers will not only hurt the healthy development of social media but also introduce many economic and social problems. Therefore, the government and enterprises must distinguish whether a hot event on microblogging platforms is a spammer or is a naturally-developing event. In this paper, we focus on the hot event list on Sina Weibo and collect the relevant microblogs of each hot event to study the detecting methods of spammers. Notably, we develop an integral feature set consisting of user profile, user behavior, and user relationships to reflect various factors affecting the detection of spammers. Then, we employ typical machine learning methods to conduct extensive experiments on detecting spammers. We use a real data set crawled from the most prominent Chinese microblogging platform, Sina Weibo, and evaluate the performance of 10 machine learning models with five sampling methods. The results in terms of various metrics show that the Random Forest model and the over-sampling method achieve the best accuracy in detecting spammers and non-spammers.

How do Mainland Chinese tourists perceive Hong Kong in turbulence? A deep learning approach to sentiment analytics

AbstractDeep learning has garnered increasing attention in many research fields. However, prior research seldom focused on tourists' perception prediction and prescription towards tourism destinations in turbulence. This study attempts to fill the gap by investigating Mainland Chinese tourists' perception of a turbulent Hong Kong society through deep learning‐based sentiment analytics. This incorporates a convolutional neural network (CNN) model for sentiment prediction and feature frequency analysis for sentiment prescription for 52,950 Chinese travel microblogs about Hong Kong. Results show that the CNN‐based deep learning approach can obtain improved sentiment predictive performance with minimal domain knowledge and human effort. The trend of Mainland Chinese tourists' (MCTs') sentiments about Hong Kong and the unique sentiment features revealed by our approach can provide practitioners with new insights into design customised tourism marketing and development strategies. The MCTs' shared mentalities emerged from sentiment features may help to enhance our theoretical knowledge about tourists' perception in turbulent tourism markets.

Chinese Microblog Sentiment Detection Based on CNN-BiGRU and Multihead Attention Mechanism

With the rapid development of the Internet, Weibo has gradually become one of the commonly used social tools in society at present. We can express our opinions on Weibo anytime and anywhere. Weibo is widely used and people can express themselves freely on it; thus, the amount of comments on Weibo has become extremely large. In order to count up the attitudes of users towards a certain event, Weibo managers often need to evaluate the position of a certain microblog in an appropriate way. In traditional position detection tasks, researchers mainly mine text semantic features through constructing feature engineering and sentiment dictionary, but it takes a large amount of manpower in feature selection and design. However, it is an effective method to analyze the sentiment state of microblog comments. Deep learning is developing in an increasingly mature direction, and the utilization of deep learning methods for sentiment detection has become increasingly popular. The application of convolutional neural networks (CNN), bidirectional GRU (BiGRU), and multihead attention mechanism- (multihead attention-) combined method CNN-BiGRU-MAttention (CBMA) to conduct Chinese microblog sentiment detection was proposed in this paper. Firstly, CNN were applied to extract local features of text vectors. Afterward, BiGRU networks were applied to extract the global features of the text to solve the problem that the single CNN cannot obtain global semantic information and the disappearance of the traditional recurrent neural network (RNN) gradient. At last, it was concluded that the CBMA algorithm is more accurate for Chinese microblog sentiment detection through a variety of algorithm experiments.

Managing image: The self-praise of celebrities on social media

Studies on self-praise have focused on ordinary people’s self-praise strategies both in oral communication and on social media. However, little research has examined how celebrities perform such speech acts. This study investigates the pragmatic strategies used in the self-praise of Chinese celebrities on social media microblogs. A dataset of 300 microblogs containing self-praise was collected from 10 Chinese celebrities (5 men and 5 women) on the Chinese microblogging site Weibo. This study analyses the types of self-praise strategies and the percentage of each strategy employed by the celebrities on Weibo. The results indicated that three main categories of self-praise strategies were employed, namely, modified explicit self-praise, implicit self-praise and explicit self-praise without modification, in decreasing order of frequency. In addition, the study explicates the factors underlying Chinese celebrities’ use of self-praise strategies to maintain a positive image. This study adds to the under-investigated areas of celebrities’ self-praise in online contexts and celebrities’ image management. Implications for politeness research in Chinese are also discussed.

Self-praise on Chinese social networking sites

Self-praise involves disclosing positive content about oneself, and such disclosures can be treated as self-elevating acts. This study adds to the growing literature on self-praise, especially in online contexts, by analysing pragmatic strategies in Chinese online self-praise from a dataset of 300 user-generated microblogs on Weibo. This study analyses the frequency and percentage of each self-praise strategy deployed by Chinese users, supplemented by a qualitative analysis of individual strategies with examples. It was found that three main pragmatic strategy categories are employed for self-praise in Chinese microblogs, namely explicit self-praise without modification, modified explicit self-praise, and implicit self-praise, in decreasing order of frequency. This study also generates hypotheses to discuss self-praising on microblogs and provides insights for research on self-praise.

Public opinion on MOOCs: sentiment and content analyses of Chinese microblogging data

ABSTRACT The increasing and widespread usage of social media enables the investigation of public preference using the web as a device. Public sentiment as expressed in 44,319 massive open online course (MOOCs) related microblogs from January to December 2017 was examined on Sina Weibo (the Chinese equivalent of Twitter) to obtain broad insight into how MOOCs are viewed by the public in the Chinese educational landscape. Despite the unstable upward trend of public interest in MOOCs over the past 12 months, the public opinion on MOOCs was largely positive. Content and sentiment analyses were conducted to facilitate a better understanding of what is communicated on social media. A general model of public opinions of MOOCs in China has been developed based on the findings. Individuals were classified into a threefold typology based on the sources and purposes of how this recent form of distance education was perceived. Based on the seven themes, the public views towards MOOCs were differentiated among ‘promoters’, ‘commenters’ and ‘experiencers’.Implications of the findings were also discussed.

Chinese Corporate Microblogging on Overseas SNS

AbstractContextualized in the “Going Global” strategy of Chinese corporations under the Belt and Road Initiative and drawing upon the attitudinal and engagement systems of appraisal, this study aims to provide semiotic and rhetorical insights into the discursive strategies of Chinese corporations in microblogging on overseas social networking sites (SNS). The posts, made over a 15-day period by the five giant Chinese companies on the US-based SNS Facebook, are analyzed both quantitatively and qualitatively. The results reveal that, in terms of attitudinal appeals, judgment resources are most preferred by Chinese corporations while affect resources are least deployed; in terms of engagement patterns, dialogic engagement patterns proliferate. It is explained that Chinese corporations strongly appeal to ethos and low power distance, demonstrating Western achievement-based cultural orientations, despite the existing discursive manifestations of the high-context, power-distanced, collectivistic characteristics of Chinese culture. This pragmatic and rhetorical shift elicits transculturality in a global cyber space informed by digital globalization and economic glocalization.

Cross-group or within-group attention flow? Exploring the amplification process among elite users and social media publics in Sina Weibo

As one of the most important forms of communicative behavior, retweeting enhances the importance or visibility of a particular object in social media. Conceptualizing retweeting as a process of amplification, this study examines how attention to a particular set of issues may be increased by the actions of verified users or social media publics on the largest Chinese microblogging website, Sina Weibo. The Granger-causality test was applied to analyze the sequential linkage between original posts and retweets by different user groups across days 1 to 14. The results show that verified users’ original posts on unobtrusive issues triggered related retweets only in their own network, while social media publics’ original posts on both unobtrusive and obtrusive issues triggered related retweets within their own group. In addition, the results demonstrate that amplification of different types of issue exhibits systematically different temporalities, that is, the response to obtrusive issues is more immediate and ephemeral than that response to unobtrusive issues. The findings highlight the possibilities and limitations of grassroots empowerment through social media in China, and provide meso-level evidence about the complex issue of attention flows in social media – whether bottom-up or top-down.

Design of Festival Sentiment Classifier Based on Social Network.

With the development of society, more and more attention has been paid to cultural festivals. In addition to the government's emphasis, the increasing consumption in festivals also proves that cultural festivals are playing increasingly important role in public life. Therefore, it is very vital to grasp the public festival sentiment. Text sentiment analysis is an important research content in the field of machine learning in recent years. However, at present, there are few studies on festival sentiment, and sentiment classifiers are also limited by domain or language. The Chinese text classifier is much less than the English version. This paper takes Sina Weibo as the text information carrier and Chinese festival microblogs as the research object. CHN-EDA is used to do Chinese text data augmentation, and then the traditional classifiers CNN, DNN, and naïve Bayes are compared to obtain a higher accuracy. The matching optimizer is selected, and relevant parameters are determined through experiments. This paper solves the problem of unbalanced Chinese sentiment data and establishes a more targeted festival text classifier. This festival sentiment classifier can collect public festival emotion effectively, which is beneficial for cultural inheritance and business decisions adjustment.

Does Length Impact Engagement? Length Limits of Posts and Microblogging Behavior

Managers of social media platforms are increasingly competing over users in order to stimulate user generated content on their sites. In this study, we delve into one tactic of this competition which has become increasingly popular, but has received limited attention, the extension of character limits. Despite its popularity, the impact of changing these limits on user behavior is uncertain. On the one hand, increasing limits might yield more low quality and rambling posts, thereby alienating the users who favor microblogs for their streamlined and efficient content. On the other hand, increasing limits might give users the freedom they need to more effectively communicate ideas. We delve into this tension by exploiting a natural experiment, the phased rollout of length limit extensions on the Chinese microblog, Weibo. Findings from more than 7,000 users suggest, as one might expect, that extending length limits increases the average length of posts. However, results indicate that the magnitude of that effect is infinitesimal (only 2 characters). Further, results indicate that extending length limits significantly increases user engagement by enticing users into posting more actively. Finally, it appears that accessing the platform through tethered or mobile connection strongly moderates these effects; inasmuch as the observed changes to post length and volumes are only observed for PC users (versus mobile-only users). This suggests a potential risk in extending length limits by shifting the platform away from mobile-only users.

Depression Risk Prediction for Chinese Microblogs via Deep-Learning Methods: Content Analysis

BackgroundDepression is a serious personal and public mental health problem. Self-reporting is the main method used to diagnose depression and to determine the severity of depression. However, it is not easy to discover patients with depression owing to feelings of shame in disclosing or discussing their mental health conditions with others. Moreover, self-reporting is time-consuming, and usually leads to missing a certain number of cases. Therefore, automatic discovery of patients with depression from other sources such as social media has been attracting increasing attention. Social media, as one of the most important daily communication systems, connects large quantities of people, including individuals with depression, and provides a channel to discover patients with depression. In this study, we investigated deep-learning methods for depression risk prediction using data from Chinese microblogs, which have potential to discover more patients with depression and to trace their mental health conditions.ObjectiveThe aim of this study was to explore the potential of state-of-the-art deep-learning methods on depression risk prediction from Chinese microblogs.MethodsDeep-learning methods with pretrained language representation models, including bidirectional encoder representations from transformers (BERT), robustly optimized BERT pretraining approach (RoBERTa), and generalized autoregressive pretraining for language understanding (XLNET), were investigated for depression risk prediction, and were compared with previous methods on a manually annotated benchmark dataset. Depression risk was assessed at four levels from 0 to 3, where 0, 1, 2, and 3 denote no inclination, and mild, moderate, and severe depression risk, respectively. The dataset was collected from the Chinese microblog Weibo. We also compared different deep-learning methods with pretrained language representation models in two settings: (1) publicly released pretrained language representation models, and (2) language representation models further pretrained on a large-scale unlabeled dataset collected from Weibo. Precision, recall, and F1 scores were used as performance evaluation measures.ResultsAmong the three deep-learning methods, BERT achieved the best performance with a microaveraged F1 score of 0.856. RoBERTa achieved the best performance with a macroaveraged F1 score of 0.424 on depression risk at levels 1, 2, and 3, which represents a new benchmark result on the dataset. The further pretrained language representation models demonstrated improvement over publicly released prediction models.ConclusionsWe applied deep-learning methods with pretrained language representation models to automatically predict depression risk using data from Chinese microblogs. The experimental results showed that the deep-learning methods performed better than previous methods, and have greater potential to discover patients with depression and to trace their mental health conditions.

Sentiment classification model for Chinese micro-blog comments based on key sentences extraction

With the advancements of communication technology, the growth in the number of blogs has been remarkable. Information sharing has been an interesting matter of interest for researches, as people contribute to information shared through the Internet. Such sharing is common in the Chinese language, as approximately 15% of the world’s population are native speakers of Mandarin. Due to comments shared that may contain complex sentences in such micro-blogs, the result of sentiment classification may be affected, reducing its accuracy. Aimed at this point, a Sentiment Classification model of Chinese Micro-blog Comments based on Key Sentences (SC-CMC-KS) is proposed. The key sentence extraction algorithm for Chinese Microblog Comments is presented by considering three factors to recognize the key sentences of a given comment: the sentiment attributes, location attributes, and the critical feature word attributes. Besides, a computing algorithm of sentence sentiment value that integrates both dependency relationships and multi-rules (i.e., sentence-type rule and inter-sentence rule) is designed, as well defined a modification distance describing the relationship between modification words and core words, in which the sentiment value in the phrase level is computed according to the calculation rules so that the sentiment value in sentence level is obtained based on the multi-rules. Furthermore, the sentiment classification algorithm of micro-blog comments is presented, so that the key sentences and emoticon of the complete micro-blog are weighted to compute the final sentiment value, and comments are classified according to the threshold set. Experimental results show the effectiveness of this model yet promising.

Categorization of Green Spaces for a Sustainable Environment and Smart City Architecture by Utilizing Big Data

Urban green spaces promote outdoor activities and social interaction, which make a significant contribution to the health and well-being of residents. This study presents an approach that focuses on the real spatial and temporal behavior of park visitors in different categories of green parks. We used the large dataset available from the Chinese micro-blog Sina Weibo (often simply referred to as “Weibo”) to analyze data samples, in order to describe the behavioral patterns of millions of people with access to green spaces. We select Shanghai as a case study because urban residential segregation has already taken place, which was expected to be followed by concerns of environmental sustainability. In this research, we utilized social media check-in data to measure and compare the number of visitations to different kinds of green parks. Furthermore, we divided the green spaces into different categories according to their characteristics, and our main findings were: (1) the most popular category based upon the check-in data; (2) changes in the number of visitors according to the time of day; (3) seasonal impacts on behavior in public in relation to the different categories of parks; and (4) gender-based differences. To the best of our knowledge, this is the first study carried out in Shanghai utilizing Weibo data to focus upon the categorization of green space. It is also the first to offer recommendations for planners regarding the type of facilities they should provide to residents in green spaces, and regarding the sustainability of urban environments and smart city architecture.

Fine-grained emotion classification of Chinese microblogs based on graph convolution networks

Microblogs are widely used to express people’s opinions and feelings in daily life. Sentiment analysis (SA) can timely detect personal sentiment polarities through analyzing text. Deep learning approaches have been broadly used in SA but still have not fully exploited syntax information. In this paper, we propose a syntax-based graph convolution network (GCN) model to enhance the understanding of diverse grammatical structures of Chinese microblogs. In addition, a pooling method based on percentile is proposed to improve the accuracy of the model. In experiments, for Chinese microblogs emotion classification categories including happiness, sadness, like, anger, disgust, fear, and surprise, the F-measure of our model reaches 82.32% and exceeds the state-of-the-art algorithm by 5.90%. The experimental results show that our model can effectively utilize the information of dependency parsing to improve the performance of emotion detection. What is more, we annotate a new dataset for Chinese emotion classification, which is open to other researchers.

Detecting information requirements for crisis communication from social media data: An interactive topic modeling approach

Timely and accurate information disclosure is an important step towards open and transparent governance. However, the decision on what to disclose at what time is a difficult issue because information requirements of the public are often vague and time-sensitive, particularly during a major disaster. To facilitate crisis communication, an interactive topic modeling method is proposed to extract the public information requirements and track their evolution from social media. Using a man-made disaster as an example, the proposed method is employed to analyze and present the evolution of public expression using the Chinese microblog data. First, topics from microblog data are discovered through a multi-level topic analysis. The results are interpreted into information requirements related to public demand for investigating the causes and responsible agents behind the Tianjin Port Explosion. Second, a streamgraph method is applied to visualize the topical compositions of contents and their evolution over time. The discovered patterns of information requirements perform retrospective analyses of disasters and provide a basis for concrete recommendations for government disclosing information. Our work has implication on the proactive disclosure mechanism deciding what and when to release formal investigation information to best meet the public's information requirements.

‘Hidden in plain sight’: Expressing political criticism on Chinese social media

While the proliferation of social media technologies in China has empowered the public with new opportunities for public expression and political engagement in a ‘virtual public sphere’, Chinese Internet censorship has meant that users have to develop creative ways to engage in political criticism. In a context where both mechanical and human censors are employed, Chinese users have become adept at utilizing the affordances of technology, Chinese language and cultural resources to express their opinions through social media. Drawing upon data from the Chinese microblogging website, Weibo, surrounding the major chemical explosions in 2015 in Tianjin, the study explores three discursive techniques of indirection by Chinese social media users to express political criticism in the context of censorship. The study highlights that through the creative use of quotation, allusion and irony, users challenged the authority’s official narratives of the event. The study not only demonstrates the pluralization and dynamics of Chinese online expression, but also points to a better understanding of Chinese censoring as a continuingly evolving interplay between technology and cultural forms and between layers of government and users.

Public Reactions to the Cigarette Control Regulation on a Chinese Microblogging Platform: Empirical Analysis.

BackgroundOn January 1, 2019, a new regulation on the control of smoking in public places was officially implemented in Hangzhou, China. On the day of the implementation, a large number of Chinese media reported the contents of the regulation on the microblog platform Weibo, causing a strong response from and heated discussion among netizens.ObjectiveThis study aimed to conduct a content and network analysis to examine topics and patterns in the social media response to the new regulation.MethodsWe analyzed all microblogs on Weibo that mentioned and explained the regulation in the first 8 days following the implementation. We conducted a content analysis on these microblogs and used social network visualization and descriptive statistics to identify key users and key microblogs.ResultsOf 7924 microblogs, 12.85% (1018/7924) were in support of the smoking control regulation, 84.12% (6666/7924) were neutral, and 1.31% (104/7924) were opposed to the smoking regulation control. For the negative posts, the public had doubts about the intentions of the policy, its implementation, and the regulations on electronic cigarettes. In addition, 1.72% (136/7924) were irrelevant to the smoking regulation control. Among the 1043 users who explicitly expressed their positive or negative attitude toward the policy, a large proportion of users showed supportive attitudes (956/1043, 91.66%). A total of 5 topics and 11 subtopics were identified.ConclusionsThis study used a content and network analysis to examine topics and patterns in the social media response to the new smoking regulation. We found that the number of posts with a positive attitude toward the regulation was considerably higher than that of the posts with a negative attitude toward the regulation. Our findings may assist public health policy makers to better understand the policy’s intentions, scope, and potential effects on public interest and support evidence-based public health regulations in the future.

Data Mining and Content Analysis of the Chinese Social Media Platform Weibo During the Early COVID-19 Outbreak: Retrospective Observational Infoveillance Study.

BackgroundThe coronavirus disease (COVID-19) pandemic, which began in Wuhan, China in December 2019, is rapidly spreading worldwide with over 1.9 million cases as of mid-April 2020. Infoveillance approaches using social media can help characterize disease distribution and public knowledge, attitudes, and behaviors critical to the early stages of an outbreak.ObjectiveThe aim of this study is to conduct a quantitative and qualitative assessment of Chinese social media posts originating in Wuhan City on the Chinese microblogging platform Weibo during the early stages of the COVID-19 outbreak.MethodsChinese-language messages from Wuhan were collected for 39 days between December 23, 2019, and January 30, 2020, on Weibo. For quantitative analysis, the total daily cases of COVID-19 in Wuhan were obtained from the Chinese National Health Commission, and a linear regression model was used to determine if Weibo COVID-19 posts were predictive of the number of cases reported. Qualitative content analysis and an inductive manual coding approach were used to identify parent classifications of news and user-generated COVID-19 topics.ResultsA total of 115,299 Weibo posts were collected during the study time frame consisting of an average of 2956 posts per day (minimum 0, maximum 13,587). Quantitative analysis found a positive correlation between the number of Weibo posts and the number of reported cases from Wuhan, with approximately 10 more COVID-19 cases per 40 social media posts (P<.001). This effect size was also larger than what was observed for the rest of China excluding Hubei Province (where Wuhan is the capital city) and held when comparing the number of Weibo posts to the incidence proportion of cases in Hubei Province. Qualitative analysis of 11,893 posts during the first 21 days of the study period with COVID-19-related posts uncovered four parent classifications including Weibo discussions about the causative agent of the disease, changing epidemiological characteristics of the outbreak, public reaction to outbreak control and response measures, and other topics. Generally, these themes also exhibited public uncertainty and changing knowledge and attitudes about COVID-19, including posts exhibiting both protective and higher-risk behaviors.ConclusionsThe results of this study provide initial insight into the origins of the COVID-19 outbreak based on quantitative and qualitative analysis of Chinese social media data at the initial epicenter in Wuhan City. Future studies should continue to explore the utility of social media data to predict COVID-19 disease severity, measure public reaction and behavior, and evaluate effectiveness of outbreak communication.

A Comparison of the Psycholinguistic Styles of Schizophrenia-Related Stigma and Depression-Related Stigma on Social Media: Content Analysis.

BackgroundStigma related to schizophrenia is considered to be the primary focus of antistigma campaigns. Accurate and efficient detection of stigma toward schizophrenia in mass media is essential for the development of targeted antistigma interventions at the population level.ObjectiveThe purpose of this study was to examine the psycholinguistic characteristics of schizophrenia-related stigma on social media (ie, Sina Weibo, a Chinese microblogging website), and then to explore whether schizophrenia-related stigma can be distinguished from stigma toward other mental illnesses (ie, depression-related stigma) in terms of psycholinguistic style.MethodsA total of 19,224 schizophrenia- and 15,879 depression-related Weibo posts were collected and analyzed. First, a human-based content analysis was performed on collected posts to determine whether they reflected stigma or not. Second, by using Linguistic Inquiry and Word Count software (Simplified Chinese version), a number of psycholinguistic features were automatically extracted from each post. Third, based on selected key features, four groups of classification models were established for different purposes: (a) differentiating schizophrenia-related stigma from nonstigma, (b) differentiating a certain subcategory of schizophrenia-related stigma from other subcategories, (c) differentiating schizophrenia-related stigma from depression-related stigma, and (d) differentiating a certain subcategory of schizophrenia-related stigma from the corresponding subcategory of depression-related stigma.ResultsIn total, 26.22% of schizophrenia-related posts were labeled as stigmatizing posts. The proportion of posts indicating depression-related stigma was significantly lower than that indicating schizophrenia-related stigma (χ21=2484.64, P<.001). The classification performance of the models in the four groups ranged from .71 to .92 (F measure).ConclusionsThe findings of this study have implications for the detection and reduction of stigma toward schizophrenia on social media.