Public Attention and Sentiment toward Intimate Partner Violence Based on Weibo in China: A Text Mining Approach

Abstract

The mobile internet has resulted in intimate partner violence (IPV) events not being viewed as interpersonal and private issues. Such events become public events in the social network environment. IPV has become a public health issue of widespread concern. It is a challenge to obtain systematic and detailed data using questionnaires and interviews in traditional Chinese culture, because of face-saving and the victim’s shame factors. However, online comments about specific IPV events on social media provide rich data in understanding the public’s attitudes and emotions towards IPV. By applying text mining and sentiment analysis to the field of IPV, this study involved construction of a Chinese IPV sentiment dictionary and a complete research framework. We analyzed the trends of the Chinese public’s emotional evolution concerning IPV events from the perspectives of a time series as well as geographic space and social media. The results show that the anonymity of social networks and the guiding role of opinion leaders result in traditional cultural factors such as face-saving and family shame for IPV events being no longer applicable, leading to the spiral of an anti-silence effect. Meanwhile, in the process of public emotional communication, anger often overwhelms reason, and the spiral of silence remains in effect in social media. In addition, there are offensive words used in the IPV event texts that indicate misogyny in emotional, sexual, economic and psychological abuse. Fortunately, mainstream media, as crucial opinion leaders in the social network, can have a positive role in guiding public opinion, improving people’s ability to judge the validity of network information, and formulating people’s rational behaviour.

Understanding Alcohol Use Discourse and Stigma Patterns in Perinatal Care on Twitter

“…The thematic analysis of social media content in our study is consistent with previous qualitative studies in this field of research [ 19 , 28 , 29 ]. It illustrates the ways that people talk about perinatal alcohol use, share opinions [ 52 , 53 , 54 ], and captures how Twitter users overwhelmingly deploy negative sentiment when representing the experiences of individuals (60.3% negative vs. 39.7% non-negative).…”

“…Along with these challenges, Twitter offers several advantages. Other studies have emphasized the unique outcomes from account anonymity by investigating how identity concealment has empowered domestic violence survivors to tell their stories [ 28 ] and encouraged discussions about mental health issues by providing online solidarity and inter-user support [ 29 ]. In comparison to Facebook and other platforms that form social networks from known connections and require verification [ 30 ], Twitter users may be less inclined to self-regulate content.…”

The Relationship between Intimate Partner Violence and Online Help-Seeking: A Moderated Mediation Model of Emotion Dysregulation and Perceived Anonymity

“…Social media is crucial for people seeking help when experiencing IPV, and it is being used as a new way to seek help [ 6 ]. Social networking sites (SNS) have been integrated into young people’s daily lives, extending to their help-seeking behavior [ 7 ].…”

“…Therefore, with the emergence of social media, IPV has gradually changed from the private sphere to the public sphere. The high freedom and openness of social media enable the public to break through the boundaries of time and space, understand IPV events, and express their personal views directly [ 6 ]. The anonymity and interactivity of the Internet help people break through social inhibitions in real society, and so they can express their real emotions truthfully or exaggeratedly.…”

The Relationship between Mustard Import and COVID-19 Deaths: A Workflow with Cross-Country Text Mining

“…A more efficient or novel way of gaining insights from data in healthcare research is mining and analyzing online text with natural language processing (NLP) or text analytics [ 10 ]. For example, researchers of recent healthcare studies have learned public opinion and sentiment on topics such as COVID-19 vaccine boosters [ 11 ] and intimate partner violence [ 12 ] by mining textual data from social media.…”