关键词: 案例查询, 用户查询行为, 相似度

Abstract: The traditional algorithm of case search returns the related cases for users through passively responding the user search request, but it neglects the fact that the user search behavior can direct the process of the case search. This paper proposes a case search algorithm based on model of user search behavior. Through collecting the user search request, it uses the similarity between user search request to build a classifications model of user search behavior, and analyzes the classifications algorithm of user search request, and mainly discusses the process of the model of user search behavior directing the case search. The experimental results show that this method can efficiently enhance the recall rate of the retrieval results.

Key words: case search, user search behavior, similarity