Sentiment analysis and topic extraction of the twitter network of #prayforparis

Abstract

ABSTRACTSocial media includes a copious amount of sentiment‐embodied sentences. Sentiment is described as “a personal belief or judgment that is not founded on proof or certainty,” which may depict the emotional state of the user, such as happy, glad, terrified, miserable, or the author's viewpoint on a topic. In social science, emotions and sentiment make up a significant part of social life and are interconnected with social relationships. When experiencing emotions, people want to reveal those emotions to other people. This study seeks to validate whether the Emotional Contagion social theory holds true in microblogging data. This theory implies that related people tend to have more similar sentiments or opinions. Motivated by this sociological observation, the study explores the sentiment‐semantics of the Twitter network of #prayforparis through sentiment analysis and topic extraction. Social Network Analysis was conducted using NodeXL to investigate the research questions. The study implemented R for conducting sentiment analysis and generating word clouds with the collected data. The study also conducted content analysis of tweets through topic extraction by applying the most recent version of SAS Enterprise Miner (13.2). In conclusion, the results confirmed the Emotional Contagion Theory in the Twitter network of #prayforparis.