The impact of preprocessing on text classification

作者：

Highlights：

• The impact of preprocessing on text classification in terms of various aspects is extensively examined.

• Experiments are conducted on two different domains and in two different languages.

• Choosing appropriate preprocessing tasks may improve classification accuracy significantly.

摘要

•The impact of preprocessing on text classification in terms of various aspects is extensively examined.•Experiments are conducted on two different domains and in two different languages.•Choosing appropriate preprocessing tasks may improve classification accuracy significantly.

论文关键词：Pattern recognition,Text categorization,Text classification,Text preprocessing

论文评审过程：Received 27 February 2013, Revised 20 August 2013, Accepted 28 August 2013, Available online 16 September 2013.

论文官网地址：https://doi.org/10.1016/j.ipm.2013.08.006