Evaluating evidences for keyword query disambiguation in entity centric database search

Abstract

A number of existing approaches attempt to reduce ambiguity of user's keyword queries by translating them to structured database queries. This disambiguation process relies on a proper assessment of whether a structured query represents the intent behind the keyword query. In this paper we systematically analyze a number of intuitive statistical measures that can potentially be used in this disambiguation process. We evaluate the impact of these measures through experiments on real-world data. © 2010 Springer-Verlag.