GeoBurst+: Effective and real-time local event detection in geo-tagged tweet streams

Chao Zhang, Dongming Lei, Quan Yuan, Honglei Zhuang, Lance Kaplan, Shaowen Wang, Jiawei Han

Research output: Contribution to journalArticlepeer-review

Abstract

The real-time discovery of local events (e.g., protests, disasters) has been widely recognized as a fundamental socioeconomic task. Recent studies have demonstrated that the geo-tagged tweet stream serves as an unprecedentedly valuable source for local event detection. Nevertheless, how to effectively extract local events from massive geo-tagged tweet streams in real time remains challenging. To bridge the gap, we propose a method for effective and real-time local event detection from geo-tagged tweet streams. Our method, named GeoBurst+, first leverages a novel cross-modal authority measure to identify several pivots in the query window. Such pivots reveal different geo-topical activities and naturally attract similar tweets to form candidate events. GeoBurst+ further summarizes the continuous stream and compares the candidates against the historical summaries to pinpoint truly interesting local events. Better still, as the query window shifts, GeoBurst+ is capable of updating the event list with little time cost, thus achieving continuous monitoring of the stream. We used crowdsourcing to evaluate GeoBurst+ on two million-scale datasets and found it significantly more effective than existing methods while being orders of magnitude faster.

Original languageEnglish (US)
Article number34
JournalACM Transactions on Intelligent Systems and Technology
Volume9
Issue number3
DOIs
StatePublished - Feb 2018

Keywords

  • Data stream
  • Event detection
  • Local event
  • Location-based service
  • Social media
  • Spatiotemporal data mining

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'GeoBurst+: Effective and real-time local event detection in geo-tagged tweet streams'. Together they form a unique fingerprint.

Cite this