(英) |
We propose a visualization method which presents distinction in tones among multiple news sites on the Web. This method uses a sentiment map to take a view of the website distinction.
Tones of websites are extracted as average sentiments of the news articles which were written concerning a topic in each website. The sentiments of news articles are represented by four values calculated in four sentiment scales; "Bright , Dark," "Acceptance , Rejection,"
"Relaxation , Strain," and "Anger , Fear," and the sentiment values of news articles are calculated using the sentiment dictionary which was constructed by our previously proposed method in advance. If a user entered one or more topic keywords, our proposed system extracts
the news articles which include the keywords from each of predetermined news sites, and calculates sentiment values of the news articles and their average value in each sentiment.
The system then generates a bar graph from the four average values in each news site and attaches all the bar graphs on the map of Japan using the Yahoo Map API. We call the map "Sentiment Map" in this paper. The sentiment maps help users to intuitively and efficiently understand distinction in tones among multiple websites concerning a given topic.
In this paper, we describe how to create a sentiment map, and evaluate our proposed system through several experiments. |