Early detection method for emerging topics based on dynamic bayesian networks in micro-blogging networks

作者：

Highlights：

• We propose a new method for early detection of emerging topics in micro-blogging.

• We find two characteristics of emerging topic which influence topic diffusion.

• We build a new DBN-based model to represent the temporal evolution of keyword.

• Performance of our method leads one to two hours earlier than others.

摘要

•We propose a new method for early detection of emerging topics in micro-blogging.•We find two characteristics of emerging topic which influence topic diffusion.•We build a new DBN-based model to represent the temporal evolution of keyword.•Performance of our method leads one to two hours earlier than others.

论文关键词：Micro-blogging networks,Emerging topics,Early detection,DBNs

论文评审过程：Received 13 February 2015, Revised 29 March 2016, Accepted 30 March 2016, Available online 1 April 2016, Version of Record 11 April 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2016.03.050