一种支持语义的云环境加密文档排序检索方法

• Format: Patent
• Language: Chinese
• Published: 08.03.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

本发明公开了一种支持语义的云环境加密文档排序检索方法，包括两个阶段：第一阶段为数据预处理和外包，利用LDA模型生成文档-主题矩阵和关键词-主题矩阵，并计算主题概率向量和关键词概率，将其发送至数据检索端；生成密钥组合并对文档和文档-主题矩阵进行加密，生成加密文档集合加密文档主题矩阵发送到云服务器。第二阶段为加密文档排序检索，将生成加密形态的检索陷门，发送到云服务器端执行排序检索，返回主题语义最相关的k个加密文档；然后，将收到的加密文档进行解密，进而获得最终的明文检索结果。采用本发明可使数据预处理和排序检索的执行效率显著提高，同时也保护了检索关键词的私密性，且引入LDA模型实现了支持语义的排序检索功能。 The invention discloses a cloud environment encrypted document sorting and retrieving method supporting semantics, which comprises two stages: data preprocessing and outsourcing in the first stage, namely, generating a document-Subject Matrix and a Keyword-Subject matrix by using LDA model, calculating subject probability vector and keyword probability, and sending the subject probability vector and keyword probability to the data retrieval terminal, generating key combinations and encrypting the documents and the document-subject matrix to generate an encrypted document set and the encrypteddocument-subject matrix and sending to the cloud server; encrypted document sorting and retrieval in the second stage, na