基于搜索历史的密文检索系统研究  

Research on the Ciphertext Retrieval System Based on Search History

作　　者：谢贤明[1] 吴庆波[1] 谭郁松[1] 

机构地区：[1]国防科学技术大学计算机学院,湖南省长沙市410073

出　　处：《中国电子商情（通信市场）》2011年第4期99-104,共6页

摘　　要：数据加密是保证数据安全的有效方法。本文针对传统的密文检索方法，设计了一种基于搜索历史的密文检索系统。分别构建文档特征向量和用户兴趣向量，运用向量空间模型对其做相关性计算，返回用户感兴趣的检索结果。并且能够根据使用者的检索历史对兴趣库实现动态更新。实验表明该方法能有效提高检索结果的准确性。Abstract： Data encryption is an effective way to ensure data security. Based on the traditional ciphertext retrieval technology, in this paper we design a ciphertext retrieval system based on search history. We set up document feature vectors and user interest vector, and do their correlation calculation by using vector space model, finally return the interested results for the user. We update the users＇ interest library dynamically according to the search history. Experiments show that the method can improve the accuracy of search results effectively.

关 键 词：数据加密 密文检索 特征向量 兴趣向量 搜索历史 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]