基于层次自适应的文本分类技术的研究

s an effective technology for organizing and positioning of information, text category gained rapid development in the last decade. This paper establishes a classification based on hierarchical adaptive to the multi-level categories of data classification. At higher level, a method of randomly selection according to the number of different category is used for data homogenization. At the lower level, data is chosen proportionally. Decision Trees method is used to classify the document to get the final result. Experimental results show that the adaptive hierarchical classification performs better than simply hierarchical classification.