Obtaining Efficient Sentence Clustering Using Hierarchical Fuzzy Clustering Algorithm

Data mining can be defined as the extraction of implicit, previously unknown, and potentially useful information from data. Clustering is the process of grouping of similar data items. Sentence clustering mainly used in variety of applications such as classify and categorization of documents, automatic summary generation, organizing the documents, etc. Size of the clusters may change from one cluster to another. The traditional clustering (hard clustering) algorithms have some problems in clustering the input dataset. The problems are instability of clusters, complexity and sensitivity. To overcome the drawbacks of these clustering algorithms, this paper proposes an algorithm called Hierarchical Fuzzy Relational Eigenvector Centrality-based Clustering Algorithm (HFRECCA) which is used for the clustering of sentences. Contents present in text documents contain hierarchical structure and there are many terms present in the documents which are related to more than one theme hence HFRECCA will be useful algorithm for natural language documents.