A Two-phase Ranking Method for the Relational Databases

Abstract

With the extensive application of keyword query in relational database, the ranking algorithm has become an important research topic. Most existing ranking methods have adopted IR-style ranking models without considering the effects of semantic relevancy and diversity on the ranking results. In this paper, we propose a novel ranking method on the basis of semantic relevancy and diversity. First, we define the concepts of basic semantics and complex semantics and then add them to the ranking function. Next, aiming at the generation of redundancy, we design the diversity strategy to re-rank query results. Experimental results demonstrate that our method has better performance in terms of precision and recall.