Fuzzy Clustering of Text Documents Using Naïve Bayesian Concept

Clustering organizes text in an unsupervised fashion. In this paper, we propose an algorithm for the fuzzy clustering of text documents using the naive Bayesian concept. Fuzzy clustering implies that the text documents are assigned to multiple clusters, ranked in descending order of probability. The Vector Space Model is used to represent our dataset as a term-weight matrix. In any natural language, semantically linked terms tend to co-occur in documents. Hence, the co-occurrences of pairs of terms in the term-weight matrix are observed. This information is used to build a term-cluster matrix where each term may belong to multiple clusters. The naive Bayesian concept is used to uniquely assign each term to a single term-cluster. The documents are assigned to multiple clusters using mean computations. The proposed algorithm has been validated using benchmark datasets available on the Internet. Our results show that the proposed scheme has a significantly better running time as compared to traditional algorithms.