分析了现有检索式修正方法,结合基于文献和基于本体的检索式修正方法的优势,提出基于引用标注的检索式修正方法。通过文献间的引用关系得到候选标注词集合,研究适用于标注词推荐的加权关联规则挖掘算法,采用改进的加权关联规则挖掘算法为用户推荐标注词,利用推荐标注词修正检索式改进检索结果。 Based on the advantages of literature and ontology based retrieval method, this paper analyzes the existing retrieving method, and proposes a retrieval method based on reference mark. Through the reference relationship among documents, the candidate annotation set is obtained, and the weighted relevance rule mining algorithm for annotation recommendation is researched. The improved weighted association rule mining algorithm is used to recommend annotation words to users, and the recommended annotation is used to improve the retrieval results.