Research on the Text Data Mining and Clustering Algorithm based on the Bayesian Theory and Markov Random Fields

In this paper, we research on the text data mining and clustering algorithm based on Bayesian theory and Markov random fields. Clustering analysis is an important means of realization of text mining. However, the text data with high dimension, sparse data distribution and the core characteristics of different categories of overlapping important characteristics such as the great challenges for practical applications. In recent years, the soft subspace clustering technology has become the focus in the academic circles. Its basic idea is to give the data set of all kinds of don't give the characteristics of the different weight vectors, used to represent the clustering process in the contribution of each feature in this category. Our algorithm combines the advances of the Bayesian theory and Markov random fields to help enhance the overall performance.