An adaptive topic tracking method based on feedback stories
Yan Ling Zheng, Ran Lu · 2012
In topic tracking, topic model influences the accurate of topic tracking, and the features of topic tracking that initial topic related stories are few and topic evolutes dynamically, lead to initial topic model could not express a topic accurately. So, in this paper, we proposed an adaptive topic tracking method based on feedback stories, amend topic model dynamically by feedback stories, which collected through dynamic threshold, and increase the weight of named entity to express a topic better. Experimental results indicate that, this method could solve the topic shifting problem effectively, and the miss tracking rate and fault tracking rate decreased a lot in topic tracking.