Optimization of Query Expansion Source in Formal Concept Analysis

In this paper, we propose a novel, hybrid approach to optimize the query expansion source. The main idea is that, on one hand, we analysis the documents selected by user and use the knowledge of Formal Concept Analysis to construct a user concept lattice; on the other hand, we find some documents which are different from the documents mentioned above and build another concept lattice. Then we calculate the concept similarity through a novel method to identify the concepts which have higher similarity with the user concept lattice. Finally, we extract the extension of these concepts and add them to the user document set; consequently we can optimize the query expansion source. The experiment results show that our method is quite reasonable.