A Method of Text Dimension Reduction Based on CHI and TF-IDF

DOI

10.2991/icmmcce-15.2015.356How to use a DOI?

Keywords

Dimension extraction; CHI; TF-IDF; Text classification

Abstract

In order to improve the result of text dimension extraction, a text dimension reduction method which is based on CHI and TF-IDF is designed and realized. News websites on the Internet provide news articles, through the page analysis based on SVM and application method designed based on CHI and TF-IDF, good results achieved by extracting text dimensions from news website. According to the news articles from NetEase and ChinaNews, the proposed algorithm model of the text dimension extraction is designed and achieved 81.2% accuracy of the text classification, which provided the data support for the method designed. This method can make up a low frequency word defect of CHI and get efficient result on text classification.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - HaiBo Tang
AU  - Lei Zhou
AU  - Xu Chengjie
AU  - Quanyin Zhu
PY  - 2015/12
DA  - 2015/12
TI  - A Method of Text Dimension Reduction Based on CHI and TF-IDF
BT  - Proceedings of the 4th International Conference on Mechatronics, Materials, Chemistry and Computer Engineering 2015
PB  - Atlantis Press
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmmcce-15.2015.356
DO  - 10.2991/icmmcce-15.2015.356
ID  - Tang2015/12
ER  -