RANKING LOCATION-DEPENDENT KEYWORDS FROM MICROBLOGS
Satoshi Ikeda, Nobuharu Kami, Takashi Yoshikawa
2012
Abstract
The spread of microblogging services, such as Twitter, has made it possible to extract location-dependent context such as keywords specific to a geographical region, with fine granularity. The results of content analysis of microblogging services are affected by users who post excessive messages. In addition, because geographical granularity of users’ interests differs, it is preferable to support multiple levels of granularity for usability. Thus, we propose a ranking method of location-dependent keywords based on a term frequency-inverse document frequency method, which takes into account diversity of information sources and supports multiple zoom levels of geographical areas by approximation. We evaluated our ranking method with a real dataset from Twitter and showed its effectiveness. We also describe a prototype implementation of a system using our ranking method.
DownloadPaper Citation
in Harvard Style
Ikeda S., Kami N. and Yoshikawa T. (2012). RANKING LOCATION-DEPENDENT KEYWORDS FROM MICROBLOGS . In Proceedings of the 8th International Conference on Web Information Systems and Technologies - Volume 1: WEBIST, ISBN 978-989-8565-08-2, pages 607-615. DOI: 10.5220/0003905206070615
in Bibtex Style
@conference{webist12,
author={Satoshi Ikeda and Nobuharu Kami and Takashi Yoshikawa},
title={RANKING LOCATION-DEPENDENT KEYWORDS FROM MICROBLOGS},
booktitle={Proceedings of the 8th International Conference on Web Information Systems and Technologies - Volume 1: WEBIST,},
year={2012},
pages={607-615},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003905206070615},
isbn={978-989-8565-08-2},
}
in EndNote Style
TY - CONF
JO - Proceedings of the 8th International Conference on Web Information Systems and Technologies - Volume 1: WEBIST,
TI - RANKING LOCATION-DEPENDENT KEYWORDS FROM MICROBLOGS
SN - 978-989-8565-08-2
AU - Ikeda S.
AU - Kami N.
AU - Yoshikawa T.
PY - 2012
SP - 607
EP - 615
DO - 10.5220/0003905206070615