Understanding Sentiment of People from News Articles: Temporal Sentiment Analysis of Social Events
Tomohiro Fukuhara, Hiroshi Nakagawa and Toyoaki Nishida
Temporal sentiment analysis that analyzes temporal trends of sentiments and topics from a text archive that has timestamps is proposed. The method accepts texts with timestamp such as Weblog and news articles, and produces two kinds of graphs, i.e., (1) topic graph that shows temporal change of topics associated with a sentiment, and (2) sentiment graph that shows temporal change of sentiments associated with a topic. Sample results obtained by applying the method to news articles are described.
Poster
Available as PDF