A preliminary approach to detect and track events in social media

Many algorithms have been proposed to model spatiotemporal events in both sensor network and social networks. However, most of them can not fullfil the task in a social network data streaming context. We proposed an evolving Mean Shift clustering based algorithm to formulate a robust system to automatically detect and track events in social network media. We also demonstrate its performance in empirical experiments. Our online system can be udapted and maintained without comsuming too much system resources which may formulate a good basis for event detection and tracking in the domain of real-time social network media.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2016

Author dc:creator

• Tang, Minyi

Contributors dc:contributor

• Abdelzaher, Tarek F.

• Social Event Tracking
• Data Mining

Statement dc:rights

• Copyright 2016 Minyi Tang

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/90843

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/90843