Automatic Web Document Classification Based on Category Axis of Terms with Increased Weight
Zhi‐Qiang Zhang, Zheng Jia-heng · Jisuanji gongcheng · 2004
Based on the analyses of traditional Nearest Neighbor text categorization and features of Web documents, a new method is put forward to effectively organize the rich information on the Internet. It has the features of drawing on the structural information of Web documents to increase the weight of the distinctive terms and constructing the categorization model on the basis of category axis vector. The experiment shows that this method has achieved a satisfactory result in both precision and recall.