Dynamic Ontology Based Model for Text Classification

Text Mining is the process of extraction of knowledge from large amounts of text. Information retrieval gains more focus due to tremendous increase of textual information in web servers. It is a very challenging task to process, organize, analyze and extract knowledge from huge volumes of unstructured text. Traditional text classification algorithms needs well defined text corpus to train and classify the given textual information. It is a very complex task to build text corpus with the help of thesaurus on the given text documents. To overcome this, ontology models can be considered for constructing knowledge base. Domain ontologies are considered for extracting more useful and high quality data on a particular domain. Many traditional approaches use static methods to construct domain ontologies. But, ontologies developed by these static approaches consist of limited terms in their knowledge base due to lack of updation. In this paper, we propose a dynamic ontology based model to classify the extracted terms and to build knowledge base on a particular domain. Experimental results show that our dynamic ontology based model performing excellently by frequent updating of extracted terms in our knowledgebase.