DFSSM based Web text clustering algorithm

A key challenge of data mining is to tackling the problem of mining richly structured datasets such as Web pages. In this paper, we propose a Web text clustering algorithm (WTCA) based on DFSSM, which is our original work. The algorithm includes the training stage of SOM and the clustering stage. It can distinguish the most meaningful features from the Concept Space without the evaluation function. We applied the algorithm to the Chinese Modern Long-distance Education Network, and compared our work with some popular clustering algorithms. The experimental results show that the average accuracy of WTCA is better than that of the other three algorithms.