WordNet and Wiki Based Approach for Finding Polysemy Tags in a Tag Set

The tagging process involves use of labels to provide content with further information using a set of keywords (tags). A tag is a keyword or phrase used to provide metadata about a resource. Resource can be image, web page, video, audio etc., which are available on Web. In this research article the issue of polysemy (a word or phrase with multiple meanings) in tagging is analyzed because it creates a lot of confusion and is cause of ambiguity, redundancy and incorrect search results. In particular if a user looks at URLs the user can’t guess about what sort of resource it refers. By looking at tag set associated with a resource a user can guess about the resource, but the guess can be ambiguous if tag set contains polysemy tags. A novel solution is proposed for the detection of polysemy tags in folk tag set. The capabilities of Word Net, Wikipedia and visual tag dictionary are exploited. Experiments are performed on tag sets that is taken from Delicious. This approach can detect many common types of polysemy that appears in folksonomies like contrastive and complementary polysemy.