Pinnacle SMU LARC

TwiCube: A Real-time Twitter Off-line Community Analysis Tool.

Overview

As a micro-blogging service, Twitter differs from other social network services in two ways: 1) the absence of mutual consent in establishing follow links and 2) being a mixture of news media and social network. A key question to ask in better understanding Twitter user behavior is which part of a user's Twitter network reflects one's real-life social network. TwiCube is an online tool that employs a novel algorithm capable of identifying a user's real-life social community, which we call the user's off-line community, purely from examining the link structure among the user's followers and followees. Based on the identified off-line community, TwiCube provides a summary of the user's interests, tweeting habits and neighborhood popularity analysis. Evaluations from real Twitter users demonstrate that our off-line community detection approach achieves high precision and recall in most cases.

Members

Juan DU, Wei XIE, Feida ZHU

Dataset

Twitter, Sina Weibo

Demo

TwiCube

Publication

"TwiCube: A Real-time Twitter Online Community Analysis Tool"
Juan Du, Wei Xie, Cheng Li, Feida Zhu and Ee-Peng Lim, The 18th International Conference on Database Systems for Advanced Applications (DASFAA'13), Wuhan, China, April, 2013.