An Algorithm of Semi-supervised Web-Page Classification Based on Fuzzy Clustering
Geng Chen, Zhu Yuquan, Jianing Tan, Hu Tianhan · 2009
It is very difficult to obtain labeled training samples. However, it is very easy to obtain non-labeled training samples. So,it is important task that how to classify Web-page using these training samples. An Algorithm called FC-TSVM based on fuzzy clustering is proposed. The algorithm FC-TSVM uses the fuzzy clustering algorithm to determine the number of positive label samples, and add the information of homepages hyperlink as part of the classifications. The experiments show that the algorithm FC-TSVM can efficiently improve the accuracy and stability of web page classification.