Topic structure mining using temporal co-occurrence

Hiroyuki Toda, Hiroyuki Kitagawa, Ko Fujimura, Ryoji Kataoka
2008 Proceedings of the 2nd international conference on Ubiquitous information management and communication - ICUIMC '08  
This paper proposes a topic structure mining method for document set with timestamp. The topic structure mining is a text mining method using the graph structure based on document pair similarities in the document set. This method yields not only topic extraction from documents and clustering of documents but also extracting the relationship between clusters and meaning of each document in the cluster. Our method combines the temporal co-occurrance and document similarity to construct the graph
more » ... construct the graph structure. We also report the evaluation result and effect of this proposed method.
doi:10.1145/1352793.1352843 dblp:conf/icuimc/TodaKFK08 fatcat:dfaqbaymsfaehl5abbix36gbsm