Concept-based information access

In recent years, the available information on the Internet is growing at a very fast rate. As the volume of information in the web getting bigger, finding relevant information in such an unstructured environment is becoming a difficult problem. Current keyword matching does not suffice since it lacks the conceptual matching of user query and information resources. Since words can be ambiguous, the search will give many irrelevant results. In order to solve this problem, we need to access information resources based on concepts. In order to go from words to concept level, we apply Word Sense Disambiguation in order to obtain intended concepts or meaning. Currently, WordNet electronic thesaurus is used for this purpose. Concepts are defined as senses in WordNet. In addition to concept identification based on WSD, we also try another method to find concepts in a domain specific text assuming that only one sense of a word is relevant for the domain.