Journal of Theoretical and Applied Information Technology
Text clustering methods have been discussed with various factors. The authors have the goal and problem of making decision with rationality factors about how much information is available or how complete the document about the knowledge. They propose a new rational text clustering algorithm using the semantic ontology. The documents are processed to extract the key terms as feature vectors. Semantic Frequency (SF) and Inverse Semantic Frequency (ISF) are computed for each document. Using computed SF and ISF they compute semantic weight for each document towards various categories, based on which the document is identified to a class or category.