Web Text Categorization Based on Latent Semantic Analysis

摘要

Traditional text categorization methods are difficult to deal with the high dimensionality characteristic of the text document based on semantic concept of the words.This paper proposed a method to Web text categorization based on latent semantic analysis. Textual data was mapped into a lower space. The proposed approach used the singular-value decomposition to derive a latent semantic space.The SVM is used to text categorization in the semantic space.Experimental results show that the method is effective on the performance of the text categorization.