A Representation Model based on Wordnet Meanings for Internet Search

Andrés Carretero, Francisco P. Romero, José A. Olivas

The majority of search engines obviate the different meanings of the words used in the searching process. Introducing meanings in this process could improve the results, but the use of wrong meanings could produce worse results. To deal with this problem, in this work is presented a document representation model based on meanings. In this model is developed an automatic word sense disambiguation process. Index terms are associated with their corresponding meanings in WordNet. The model proposed is implemented in an experimental search engine called Bianca which has been tested using standard measures. The achieved results are compared with a classic word searcher.

PDF full paper