云计算下支持语义的可搜索加密方法研究

传统的云计算下的可搜索加密算法没有对查询关键词进行语义扩展,导致了用户查询意图与返回结果存在语义偏差,并且对检索结果的相关度排序不够合理,无法满足用户对智能搜索的需求。对此,提出了一种支持语义的可搜索加密方法。该方法利用本体知识库实现了用户查询的语义拓展,并通过语义相似度来控制扩展词的个数,防止因拓展词过多影响检索的精确度。同时,该方法利用文档向量、查询向量分块技术构造出对应的标记向量,以过滤无关文档,并在查询-文档的相似度得分中引入了语义相似度、关键词位置加权评分及关键词-文档相关度等影响因子,实现了检索结果的有效排序。实验结果表明,该方法在提高检索效率的基础上显著改善了检索结果的排序效果,提高了用户满意度。

Research on Searchable Encryption Method with Supporting Semantics in Cloud Computing

The traditional searchable encryption(SE)algorithm in cloud computing doesn’t extend the query keywords semantically,which leads to the semantic deviation between the query intention and the returned result,and the inordinate relevance of the search results is not reasonable enough to meet the needs of users for intelligent search.In this study,a SE method of supporting semantics is proposed.The ontology knowledge base is used to realize the semantic extension of user query,and the number of extended words by semantic similarity is seriously controlled to improve the retrieval accuracy without overextending words.Furthermore,by using document vector and query vector block technology,the corresponding marker vector is constructed to filter irrelevant documents,and semantic similarity,keyword location weighted score and keyword-document relevancy are introduced into query-document similarity score,so as to effectively sort the retrieval results.The experiment shows that the proposed method can significantly increase the retrieval efficiency and improve the user’s satisfaction.