Digital Library

cab1

 
Title:      TOPIC-BASED BROWSING OF TENDENCIES IN TWITTER CONVERSATIONS BY USING SOCIAL GLOCAL INFORMATION EXTRACTION
Author(s):      Takafumi Nakanishi, Kiyotaka Uchimoto, Yutaka Kidawara
ISBN:      978-972-8939-82-3
Editors:      Piet Kommers and Pedro Isaías
Year:      2013
Edition:      Single
Keywords:      Topic-Based Browsing, Content Analysis, Social Glocal Information Extraction (sGIE), Micro-blog, Twitter, Tendency of Conversation Topic.
Type:      Full Paper
First Page:      316
Last Page:      325
Language:      English
Cover:      cover          
Full Contents:      click to dowload Download
Paper Abstract:      In this paper, we present a topic-based browsing method of the tendencies in micro-blog conversations. Micro-blogs such as twitter are used by various users for not only communication with friends but also extraction of important news in local areas, serving in effect as human sensors. We can see situations in more detail than with conventional media such as TV and radio. Recently, some researchers have been working on extraction methods of feature words or trajectories of focused words in the spatiotemporal axis. Twitter itself also provides a “Trend” service that identifies hot topic words. Using these services, it might be possible to find topics of focus on Twitter. However, there is no way to tell why certain words attract attention, or what becomes hot due to the words used, by simply extracting feature words and word trajectories. We are working on a new word-extraction method for social media: Social Glocal Information Extraction (sGIE). sGIE extracts important elements by comparing a global set with a local set, and it discovers the tendencies extracted from tweets, including the focal topic word. By using our method applying sGIE, we can see the tendencies in our society. In other words, sGIE provides perspectives on the social events extracted from Twitter. Accordingly, sGIE dynamically gives a schema for the focused topic after it is posted. In addition, we present a new browsing method for Twitter: Topic-Based Browsing of Conversation Tendencies in Twitter. We also demonstrate implementation of this method and show examples of browsing.
   

Social Media Links

Search

Login