Enhanced Heartbeat Graph for emerging event detection on Twitter using time series networks

作者：

Highlights：

• Burstiness based approaches ignore minor details important for event detection.

• Proposed approach captures divergence in a Twitter stream to detect events.

• Divergence is captured through topological and sequential relations using graphs.

• Proposed approach uses three novel features to detect emerging events.

• The proposed approach is efficient and outperforms state-of-the-art methods.

摘要

•Burstiness based approaches ignore minor details important for event detection.•Proposed approach captures divergence in a Twitter stream to detect events.•Divergence is captured through topological and sequential relations using graphs.•Proposed approach uses three novel features to detect emerging events.•The proposed approach is efficient and outperforms state-of-the-art methods.

论文关键词：Event detection,Twitter,Text stream,Emerging trends,Dynamic graph,Time series network,Big data

论文评审过程：Received 8 October 2018, Revised 9 April 2019, Accepted 3 June 2019, Available online 11 June 2019, Version of Record 24 June 2019.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.06.005