Use of noun phrases in identification of a website

In this paper a methodology is proposed to prepare a domain ontology of a given website. For the same we parse through the contents of website for noun pharses. We apply Term Frequency and Inverse Document Frequency (TF-IDF). This gives us the required information about the domain which can be used to build an ontology. The ontology thus created will represent the proper taxonomy of classes and their interrelationships. This ontology will help in identification of topics in website and ability to give relevant output when complex queries are given. The system will then make use of the ontology to classify it in comparison with the ontologies in the database. The system is tested on an academic instritutes website, namely, College of Engineering Pune (COEP) and proves to be useful.