Clustering Data Text Based on Semantic

Clustering is one of the most important data mining techniques which categorize a large number of unordered text documents into meaningful and coherent clusters. Most of text clustering algorithms do not consider the semantic relationships between words and do not have the ability to recognize and use the semantic concepts.In this paper, a new algorithm has been presented to cluster texts based on meanings of the words. First, a new method has been presented to find semantic relationship between words based on Wordnet ontology then, text data is clustered using the proposed method and hierarchical clustering algorithm. Documents are preprocessed, converted to vector space model, and then are clustered using the proposed algorithm semantically. The experimental results show that the quality and accuracy of the proposed algorithm are more reliable than the existing hierarchical clustering algorithms.