A Verifiable Semantic Searching Scheme by Optimal Matching over Encrypted Data in Public Cloud

Abstract of the project:

Semantic searching over encrypted data is a crucial task for secure information retrieval in public cloud. It aims to provide retrieval service to arbitrary words so that queries and search results are flexible. In existing semantic searching schemes, the verifiable searching does not be supported since it is dependent on the forecasted results from predefined keywords to verify the search results from cloud, and the queries are expanded on plaintext and the exact matching is performed by the extended semantically words with predefined keywords, which limits their accuracy. In this paper, we propose a secure verifiable semantic searching scheme. For semantic optimal matching on ciphertext, we formulate word transportation (WT) problem to calculate the minimum word transportation cost (MWTC) as the similarity between queries and documents, and propose a secure transformation to transform WT problems into random linear programming (LP) problems to obtain the encrypted MWTC. For verifiability, we explore the duality theorem of LP to design a verification mechanism using the intermediate data produced in matching process to verify the correctness of search results. Security analysis demonstrates that our scheme can guarantee verifiability and confidentiality. Experimental results on two datasets show our scheme has higher accuracy than other schemes. Index Terms—public cloud, results verifiable searching, secure semantic searching, word transportation

Existing System:

Most of the existing secure semantic searching schemes consider the semantic relationship among words to perform query expansion on the plaintext, then still use the query words and extended semantically related words to perform exact matching with the specific keywords in outsourced documents. We can roughly divide these schemes into three categories: secure semantic searching based synonym secure semantic searching based mutual information model secure semantic searching based concept hierarchy. We can see that these schemes only use the elementary semantic information among words. Introduce the Word2vec technique to utilize the semantic information of word embeddings, their approach damages the semantic information due to straightly aggregating all the word vectors. We think that secure semantic searching schemes should further utilize a wealth of semantic information among words and perform optimal matching on the ciphertext for high search accuracy.

Proposed System:

In this paper, we propose a secure verifiable semantic searching scheme that treats matching between queries and documents as an optimal matching task. We treat the document words as “suppliers,” the query words as “consumers,” and the semantic information as “product,” and design the minimum word transportation cost (MWTC) as the similarity metric between queries and documents. Therefore, we introduce word embeddings to represent words and compute Euclidean distance as the similarity distance between words, then formulate the word transportation (WT) problems based on the word embeddings representation. However, the cloud server could learn sensitive information in the WT problems, such as the similarity between words. For semantic optimal matching on the ciphertext, we further propose a secure transformation to transform WT problems into random linear programming (LP) problems. In this way, the cloud can leverage any readymade optimizer to solve the RLP problems and obtain the encrypted MWTC as measurements without learning sensitive information. Considering the cloud server may be dishonest to return wrong/forged search results, we explore the duality theorem of linear programming (LP) and derive a set of necessary and sufficient conditions that the intermediate data produced in the matching process must satisfy. Thus, we can verify whether the cloud solves correctly RLP problems and further confirm the correctness of search results. Our new ideas are summarized as follows:

• Treating the matching between queries and documents as an optimal matching task, we explore the fundamental theorems of linear programming (LP) to propose a secure verifiable semantic searching scheme that performs semantic optimal matching on the ciphertext.
• For secure semantic optimal matching on the ciphertext, we formulate the word transportation (WT) problem and propose a secure transformation technique to transform WT problems into random linear programming (LP) problems for obtaining the encrypted minimum word transportation cost as measurements between queries and documents.
• For supporting verifiable searching, we explore the duality theorem of LP and present a novel insight that using the intermediate data produced in the matching process as proof to verify the correctness of search results.

System Architecture and modules of the project:

• DATA OWNER Data Owner has to register in the cloud then he can able to upload data into cloud with encrypted. The encrypted data he can able to view his own data as well he can able to see others encrypted data also. He can able to view his file request given by the data user. He can able to give the permission to the user.He can able to view the top-k files in the cloud.
• DATA USER Data User has to register then he can able to login into the cloud .he can able to to search the data in cloud uploaded by the data owners. He can able to sent the file request to the data owner. After getting the file download permission he can able to download from the cloud.
• CLOUD SERVER. Cloud server can able to login into application. Then cloud can able view data owner and data user details. He can able to view all uploaded files by the data owners. All file requests also he can able to view.Cloud can able to view the total files as well as top-k files.