Optimization for Vietnamese text classification problem by reducing features set

摘要

Vietnamese is the single syllable language, so that process of word segmentation is relatively complex, if split word based on whitespaces, it is not accuracy, on the other hand Vietnamese segmentation tools are not high effective. In this paper, we propose a new method that used only topic word for calculating to increase accuracy of the Vietnameses text classification system and optimize the process of calculating. The experimental results show that our method more effective than the proposed approach, higher accuracy and reduce the computational complexity.