Efficient and effective ranking in Top-k exploration for keyword search on RDF

Ranking solutions is an important issue in Information Retrieval because it greatly influences the quality of results. In this context, keyword based search approaches use to consider solutions sorting as least step of the overall process. Ranking and building solutions are completely separate steps running autonomously. This may penalize the retrieving information process because it binds to order all found matching elements including (possible) irrelevant information. In this paper we present a theoretical study of YAANII, a novel technique to keyword based search over semantic data. We demonstrate how effectiveness of the answers depends not so much on the quality of the scoring metrics but on the way such criteria are involved.