A Robust Document Proximity Based Approach and Location Aware Keyword Query Suggestion in Web

Abstract

Keyword suggestion in web search helps user to access relevant information without having to known how to precisely express their queries Exiting keyword suggestion techniques do not consider the location of user and the query result the spatial proximity of user to the retrieved result is not taken as a factored in the recommendation. However the relevance’s of search result in many application location based services is known to be correlated with proximity to the query issuer. Each query is related to one of topics identified in the conversion fragments preceding the recommendation and is submitted to a search engine over the English we propose in this paper an algorithm foe diverse merging of these lists using a sub modular reward function that reward the topical similar of documents to the conversation words as well as there diversity. We evaluates the proposed method through crowd sourcing the result superiority of the diverse merging technique over Several other which enforce the diversity of topics