Feature Extraction in Text Clustering Based on Theme

Abstract

A new method is proposed, which refers to feature extraction based on oil theme of concept hierarchy to improve the weights between the high-frequency words and low-frequency words in the documents, and we use hash technology to improve the limitations of the theme of concept hierarchy. The method can identify the theme of texts accurately, and enhance the characteristic expression of texts. To a certain extent, it has resolved the semantic problem in specific areas.

2 Figures and Tables

Cite this paper

@article{Shi2008FeatureEI, title={Feature Extraction in Text Clustering Based on Theme}, author={Nianyun Shi and Kong Jing and Jiuyun Xu and Yongxiang Duan and Chunhua Li}, journal={2008 International Symposium on Intelligent Information Technology Application Workshops}, year={2008}, pages={632-635} }