基于向量空间模型的短文本密文检索方法  

Ciphertext retrieval method of short text based on vector space model

作　　者：向广利[1] 陶然[1] 林香 禹璐 XIANG Guang-li TAO Ran LIN Xiang YU Lu(College of Computer Science and Technology, Wuhan University of Technology, Wuhan 430070,Chin)

机构地区：[1]武汉理工大学计算机科学与技术学院,湖北武汉430070

出　　处：《计算机工程与设计》2017年第11期2909-2913,共5页Computer Engineering and Design

摘　　要：为提高用户在云服务等第三方存储机构的短文本的检索效率,通过对索引结构的改进,提出"关键词-主题"索引和"主题-文本"索引,结合向量空间模型的特点,使用短文本主题构建算法,提出基于时间和主题的向量空间模型,实现关键词向量维度的降低,提高短文本检索速度。实验结果表明,该向量空间模型是安全有效的。To improve the efficiency of short text retrieval in the cloud services and other third-party storage institutions,the structure of traditional index was improved,and the indexes of Keyword-Theme and Theme-Text were proposed.Combined with the characteristics of vector space model,short text theme construction algorithm was used,and the vector space model based on time and theme was proposed.Thus the method reduced the dimension of keyword vector and speeded up short text retrieval.Experimental results show that the vector space model is safe and effective.

关 键 词：短文本 密文检索 关键词 主题 向量空间模型 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]