Text Clustering Research on the Max Term Contribution Dimension Reduction and Simulated Annealing Algorithm
最大词重降维算法与模拟退火算法相结合的文本聚类方法研究

This paper presents a new algorithm for text character extraction and dimension reduction based on the Max Term Contribution. Its main idea is computing the contribution of each term in the high dimension document-base and extracting the maximum contribution terms to construct a low dimension document-base from the high dimension document-base using the search algorithm. Then a modified K-means clustering method based on the Simulated Annealing (SA) is presented to cluster the low dimension document datum which is obtained by MTC. Finally, some experiments show that the new method can improve the cluster precision.