一种基于情感计算与层次化多头注意力机制的负面新闻识别方法

网络负面新闻识别在网络舆情监测中具有重要的研究意义.针对当前海量数据下负面新闻难以检测的问题,提出了一种基于情感计算与层次化多头注意力机制相结合的负面新闻识别方法.首先,从新闻文本中采用TF-IDF（Term Frequeney-Inverse Document Frquency）和语义相似度算法构建负面新闻情感词库;其次,采用情感倾向计算方法计算负面新闻情感词的情感倾向度;最后,将词语和词语的情感倾向度进行向量化表示,并采用层次化多头注意力机制进行正负面新闻的判定.情感计算和多头注意力机制的引入,对于捕获文本中的情感词语提供了很大帮助.最终本文基于真实的网络新闻文本数据与现有的多种算法进行对比,证明了该模型具有较好的识别效果,相比于Han模型和LSTM模型分别提高了0.67%和3.29%.     

Fewer people,more flames: How pre-existing beliefs and volume of negative comments impact online news readers’ verbal aggression

In recent years, scholars have documented an increase in flaming behavior, which refers to a strong and hostile expression of emotions and feelings online. In order to identify the mechanism and specific circumstances of flaming, this study employed emotional contagion as the main theoretical framework and conducted a 2 (argument presented in a news article: support gun control vs. against gun control) × 2 (volume of negative comments: small vs. large) × 2 (news readers’ pre-existing beliefs: support gun control vs. against gun control) between-subjects experiment. Participants were asked to read an online news article about a gun-related topic as well as negative reader comments before they wrote a comment of their own. The experimental results showed that participants were more likely to use swear words in their comments when their viewpoint was diametrically opposed to the argument presented in the news article and when the article was accompanied by a low volume of negative comments. This interaction effect was mediated by negative emotions that participants experienced when reading the article and its associated comments. These findings have theoretical implications for future research pertaining to the identification of online circumstances and causes that could help mitigate the occurrence of flaming.     

Analysis of Physiological Responses and Use of Fuzzy Information Granulation–Based Neural Network for Recognition of Three Emotions

In this study, we investigate the relationship between emotions and the physiological responses, with emotion recognition, using the proposed fuzzy information granulation–based neural network (FIGNN) for boredom, pain, and surprise emotions. For an analysis of the physiological responses, three emotions are induced through emotional stimuli, and the physiological signals are obtained from the evoked emotions. To recognize the emotions, we design an FIGNN recognizer and deal with the feature selection through an analysis of the physiological signals. The proposed method is accomplished in premise, consequence, and aggregation design phases. The premise phase takes information granulation using fuzzy c‐means clustering, the consequence phase adopts a polynomial function, and the aggregation phase resorts to a general fuzzy inference. Experiments show that a suitable methodology and a substantial reduction of the feature space can be accomplished, and that the proposed FIGNN has a high recognition accuracy for the three emotions using physiological signals.     

Fear and Anger Responses to Local News Coverage of Alcohol‐Related Crimes,Accidents, and Injuries: Explaining News Effects on Policy Support Using a Representative Sample of Messages and People

An experiment investigated emotional reactions to news on policy support. Stimuli were selected from a nationally representative sample of local crime/accident news, and a nationally representative online panel of U.S. adults. Stories were manipulated to mention or not mention the role of alcohol. Anger elicited by stories increased blame of individuals, whereas fear increased consideration of contributing societal factors. Mention of alcohol increased likelihood of different emotional responses dominating—greater anger when alcohol was mentioned and greater fear when not mentioned. Such emotions influence policy support: enforcement of existing laws controlling individual behavior in addition to new laws when anger predominated, and, indirectly, support for new laws changing social context in which alcohol is promoted and sold when fear predominated.     

The Social News Gap: How News Reading and News Sharing Diverge

This article seeks to explain variation in news sharing patterns on social media. It finds that news editors have considerable power to shape the social media agenda through the use of “story importance cues” but also shows that there are some areas of news reporting (such as those related to crime and disasters) where this power does not apply. This highlights the existence of a social “news gap” where social media filters out certain types of news, producing a social media news agenda which has important differences from its traditional counterpart. The discussion suggests that this may be consequential for perceptions of crime and engagement with politics; it might even stimulate a partial reversal of the tabloidization of news outlets.     

Storytelling for Social Change: Leveraging Documentary and Comedy for Public Engagement in Global Poverty

Narrative is essential for public engagement with global poverty. Stand Up Planet, a documentary about global development, was produced to evaluate the effects of a little-utilized nonfiction comedy narrative. Using a pretest–posttest experimental design, this study examines shifts in U.S. audience engagement with global poverty after watching Stand Up Planet, compared with a somber documentary, The End Game. Both documentaries increased awareness of global poverty, support for government aid, knowledge, and intended actions. However, Stand Up Planet produced significantly larger gains in awareness, knowledge, and actions; these effects were mediated by the narrative's relatability, positive emotions, and entertainment value. The End Game's effects were mediated by narrative transportation and negative emotions. Implications for narrative in social change campaigns are discussed.     

Moderating Effect of Group Cue While Processing News on Immigration: Is the Framing Effect a Heuristic Process?

This paper presents a study on the sociocognitive effects of news frames on immigration. One hundred and eighty‐six individuals were exposed to a newspaper story on increased immigration to Spain. The newspaper highlighted (a) the positive (economic contribution frame) versus negative (crime growth frame) consequences and (b) the group cue—Latinos versus Moroccans. In contrast with economic contribution frame, crime growth frame stimulated more negative cognitive responses toward immigration, increased the salience of immigration as a problem, generated a negative attitude toward immigration, and induced greater disagreement with positive beliefs about the consequences of immigration for the country. We conceptualized the framing effect as a heuristic process in which peripheral cues in the news story guided information processing.     

Behavioral Effects of Framing on Social Media Users: How Conflict,Economic, Human Interest,and Morality Frames Drive News Sharing

We hypothesize that generic frames influence what news people share on Facebook and Twitter through three different routes: emotions, motivations, and psychological engagement. Using a mixed-methods design, a content analysis of a representative sample of articles published in six Chilean outlets was combined with in-depth interviews with digital journalists. After controlling for issue, newsworthiness, informational utility, valence, and other confounds, results show that—across platforms—a morality frame increases news sharing, whereas a conflict frame decreases it. Emphasizing economic consequences also decreases sharing, but only on Facebook. Surprisingly, the human interest angle has no noticeable effects. These results show that news frames can have behavioral consequences, and confirm the existence of a gap between preferred frames of journalists and users.     

A Communicative Approach to Social Capital

This article advances a communicative approach to social capital that views communication as the fundamental source of societal integration. We contend that integration occurring at the system level via news consumption and at the individual level via interpersonal discussion is amplified through ties at the community level. This cross‐level interaction is theorized to encourage civic engagement, writ large, above and beyond the influences of news, talk, or social ties. This perspective distinguishes between the extent of news use and political talk and the orientation toward news consumption and political conversation. We offer evidence that communication variables, specifically news attention and exposure along with conversational frequency and orientation toward conversational understanding, interact with associational membership and network size to foster engagement.     

Understanding publics’ post-crisis social media engagement behaviors: An examination of antecedents and mediators

This study examines the psychological mechanisms underlying the process that enables publics’ individual differences (e.g., the levels of uncertainty avoidance and social media usage) to give rise to varying post-crisis social media engagement intentions (e.g., information seeking, support seeking, and negative word-of-mouth). The study confirms that this process is serially mediated by perceived threat severity, perceived susceptibility, and negative emotions. These psychological mediators, however, function differently between uncertainty avoidance and social media usage. For uncertainty avoidance, a high arousal negative emotion (e.g., anger, fear) is a more essential step to influence engagement intentions than a low arousal emotion (e.g., shame, guilt). For social media usage, however, the type of aroused negative emotions does not matter. Rather, the relationship social media usage has with information seeking is different from that with support seeking and negative word-of-mouth intentions. Social media usage induces information seeking intentions only through publics’ cognitive appraisals of the situation without the activation of negative emotions, while it induces supporting seeking and negative word-of-mouth intentions through both cognitive appraisals and negative emotion arousals.     

Please Your Self: Social Identity Effects on Selective Exposure to News About In‐ and Out‐Groups

Hypotheses were tested in a secondary data analysis of an experiment on selective exposure to positive and negative news articles about same‐ and differently aged individuals. A sample of 178 young (18–30 years) and 98 older adults (50–65 years) browsed an online news magazine while reading times were logged by software. Younger individuals (high status, high uncertainty) generally focused their reading on same‐aged individuals, with a preference on positive news about this in‐group. Older recipients (low‐status, low uncertainty) were more likely to select negative news about young individuals than positive news about this out‐group and negative news about older individuals. Furthermore, exposure to negative news about younger individuals bolstered older recipients' self‐esteem.     

广播记者的新闻敏感性与受众心理分析

作为一名广播记者,在日常工作过程中重视自身新闻敏感性建设已经成为职业的必要素养,对于个人的成长与职业进步显得十分重要。只有不断提升自身的新闻敏感性,才能够在纷繁错杂的信息世界中获取更有价值、更有现实意义的新闻报道方向,来提高新闻报道整体的实用价值,从而推动广播新闻报道的长期发展。     

How Does Insightful and Emotional Disclosure Bring Potential Health Benefits? Study Based on Online Support Groups for Women With Breast Cancer

Building upon 2 theoretical models (the cognitive adaptation model and the emotional exposure–habituation model), this research focused on 2 aspects of disclosure content—insights and emotions—and examined how women with breast cancer benefit from written disclosure in online support groups. Using survey data collected at baseline and after 4 months and messages posted in bulletin‐board‐type online groups in between, we analyzed how the content of disclosive messages predicted health outcomes. Disclosure of insights led to greater improvements in health self‐efficacy, emotional well‐being, and functional well‐being, which was mediated by lowered breast cancer concerns. Disclosure of negative emotions had no main effects on health outcomes, but weakened the unfavorable association between concerns at baseline and functional well‐being at follow‐up.     

Are News Reports of Suicide Contagious? A Stringent Test in Six U.S. Cities

Past evidence of suicidal contagion from news reports in the United States is based largely on national data prior to 1980 using proxies for suicide stories rather than local news sources. Our research examined more proximal effects of suicide news reporting for 4 months in 1993 in 6 U.S. cities controlling for a wide range of alternative sources of media and interpersonal influence. In addition, predictions for the effect based on suicide contagion theories were examined for 3 age groups (15–25, 25–44, and older than 44). Local television news was associated with increased incidence of deaths by suicide among persons younger than 25 years. Newspaper reports were associated with suicide deaths for both young persons and persons older than 44 years. An unexpected protective effect of television news reports was observed in the 25–44 age range; nevertheless, news reporting was associated with an aggregate increase in suicide deaths. The results support theories of media contagion but also suggest that media depiction can inhibit suicide among some audience members.     

Research on emotional space for movie and TV drama videos

Emotional space refers to a multi-dimensional emotional model that describes a group of subjective feelings or emotions. Since the existing discrete emotional space is mainly aimed at human’s primary emotions, it cannot describe the complex emotions evoked when watching movies. In order to solve this problem, an emotional fusion space for videos was constructed by selecting movies and TV dramas with rich emotional semantics as the research objects. Firstly, emotional words based on movie and TV drama videos are acquired and analyzed by using subjective evaluation and semantic analysis methods. Then, the emotional word vectors obtained from the above analysis are fused, reduced dimension by t-distributed stochastic neighbor embedding (t-SNE) algorithm, and clustered by bisecting K-means clustering algorithm to get a discrete emotional space for movie and TV drama videos. This emotional fusion space can obtain different categories by changing the value of the emotion classification number without re-labeling and calculation.     

Seek and You Shall Find? How Need for Orientation Moderates Knowledge Gain from Twitter Use

A Web‐based survey (N = 306) examined if Twitter use enhances individuals' news knowledge, and if so, how such effects might vary depending on their need for orientation (NFO) and the type of news (hard vs. soft). The duration of Twitter use positively predicted hard news knowledge, but only for those higher in NFO. In contrast, daily Twitter use had negative influence on high NFOs' soft news knowledge, with no corresponding effect for lows. Although high NFOs were more likely to use Twitter for information‐seeking than their less surveillance‐driven counterparts, such motivation did not facilitate knowledge gain. Systematic processing of public affairs information and selective avoidance were discussed as potential explanations for differential knowledge gain by high and low NFOs.     

Social network users clustering based on multivariate time series of emotional behavior

It is known that the social network is an excellent source for gathering the emotions of people. There are thousands of micro-blogs posted in every second and every micro-blog that may contain a variety of user＇s emotions. The users＇ collective emotional behaviors are with great impacts on today＇s societies, so it is good to find groups for society management based on users＇ emotional behavior. This article focuses on analyzing multivariate emotional behavior of users in social network and the goal is to cluster the users from a fully new perspective-emotions. The following tasks are completed： firstly, the multivariate emotion of Chinese micro-blog with vector is analyzed, and multivariate time series to describe the user＇s emotional behavior are constructed. Seconedly, considering principal component analysis （PCA） in similarity and distance similarity, the similarity of the multivariate emotion time series is measured. The contribution could be summarized as follows： groups of users though different emotions in social network are discovered. The emotional fluctuation and intensity of users are considered as well. Experiment in clustering effectively illustrates the emotional behavior characteristics of the Users in different groups.     

The Role of Identification and Perception of Just Outcome in Evoking Emotions in Narrative Persuasion

The question is addressed whether identification with a story character can evoke emotions that subsequently influence the audience's attitude. Study 1 (N = 145) manipulated identification and found it to influence the evoking of emotions as well as the audience's attitude. Study 2 (N = 115) examined whether emotions are evoked by the extent to which the story's outcome is perceived as just. The main character's culpability was manipulated thereby influencing the extent to which a bad ending was considered just. Mediation analyses attested to the greater importance of identification for evoking emotions compared to the perception of justice. The studies reveal the mediating role of emotions in narrative persuasion as well as how identification can evoke these emotions.     

Some Message Variables and the MUM Effect

A separate group of subjects scaled 55 messages with regard to pleasantness for a potential recipient (N = 53), desire to communicate (N = 55), obligation to communication (N = 57), and the probability with which they would communicate each (N = 56). These ratings were used to test the hypothesis that message transmission is a function of felt obligation and desire to transmit, and that felt desire and obligation to transmit, in turn, are functions (linear for desire; U-shaped for obligation) of message pleasantness for the recipient. Results indicated that transmission is a function of felt desire and obligation over all levels of message pleasantness (r= 97; p <.001); that the determinants of obligation and desire are qualitatively different for good news versus bad news messages, i.e., message pleasantness can account for desire (r=.85; p <.01) and obligation (r=.75; p <.01) among good news messages, but neither message pleasantness nor other inferred relevant parameters of messages, namely locus of cause of message, urgency of message, or ability to empathize with the recipient, could account for desire and obligation among bad news messages. Finally, greater variability was found for bad news messages than good news messages on rated probability of transmitting, felt obligation, and desire to transmit. This greater variability suggested greater norm ambiguity with respect to the handling of bad news than of good news messages.     

What Is Second Screening? Exploring Motivations of Second Screen Use and Its Effect on Online Political Participation

This study focuses on second screening for news, a hybrid media process that combines watching news on television and a second, web‐connected screen (i.e., smart‐phone, laptop). Based on U.S.‐national, 2‐wave panel data, the paper (a) examines people's motivations to engage in second screen use, and then (b) advances the relationship between second screening and online political behaviors. Discussing and pursuing further information are both central motivations for second screen use. Furthermore, results suggest second screening for news is a significant predictor of online political participation and a key link between TV news and political engagement as this relationship is fully mediated. The study advances theoretically and empirically ways in which an informed public opinion may partake of a more engaged democracy.