Impact of an ontology for automatic text classification

Abstract

The concept of ontologies has widely been used in various applications including email filtering and electronic news classification. It can be also used for the classification of digital documents in a library. Advancing the accuracy of classification is the main purpose of using ontologies for classification. Documents may be difficult to understand due to the vague terms used in the text. However, since ontologies represent the semantic relationships of the terms, they can be used to correctly identify the subject of a document. This study made an attempt to improve the classification accuracy of an automatic text classification system by using an ontology. Classification results given by the automatic system with and without integrating the ontology were used to evaluate the impact of the ontology for automatic classification. Results showed that 32.76% more documents and 25% more subjects were correctly classified by the ontology based system than the system prior to use of ontology.