On suffix-tree-based text clustering algorithm

Hui Shu · 2012

In order to achieve Chinese text multi-topic clustering,an algorithm based on suffix tree is proposed.The main process in English text multi-topic clustering based on suffix tree is introduced,the difference between Chinese and English is analyzed,and a suffix tree model is built up with Chinese words and phrases as units,thus,the Chinese text multi-topic clustering can be conduct after the clusters association graph is constructed.Analyses of the experimental results show that,the proposed method works accurately and quickly.

Read the paper · More papers on PaperTik