Breaking News Detection and Tracking in Twitter

Author

Phuvipadawat, Swit ; Murata, Tsuyoshi

Author_Institution

Dept. of Comput. Sci., Tokyo Inst. of Technol., Tokyo, Japan

Volume

3

fYear

2010

fDate

Aug. 31 2010-Sept. 3 2010

Firstpage

120

Lastpage

123

Abstract

Twitter has been used as one of the communication channels for spreading breaking news. We propose a method to collect, group, rank and track breaking news in Twitter. Since short length messages make similarity comparison difficult, we boost scores on proper nouns to improve the grouping results. Each group is ranked based on popularity and reliability factors. Current detection method is limited to facts part of messages. We developed an application called “Hotstream” based on the proposed method. Users can discover breaking news from the Twitter timeline. Each story is provided with the information of message originator, story development and activity chart. This provides a convenient way for people to follow breaking news and stay informed with real-time updates.

Keywords

information retrieval; social networking (online); Hotstream; Twitter; breaking news detection; breaking news tracking; popularity factor; reliability factor; short length messages; Feature extraction; Indexing; Nominations and elections; Real time systems; Reliability; Twitter; User-generated content; Information Retrieval; Real-time text-mining; Topic Detection and Tracking; Twitter;

fLanguage

English

Publisher

ieee

Conference_Titel

Web Intelligence and Intelligent Agent Technology (WI-IAT), 2010 IEEE/WIC/ACM International Conference on

Conference_Location

Toronto, ON

Print_ISBN

978-1-4244-8482-9

Electronic_ISBN

978-0-7695-4191-4

Type

conf

DOI

10.1109/WI-IAT.2010.205

Filename

5616930