Research On Sorting Retrieval Scheme Of Unstructured Data Ciphertext

A large number of documents are transferred and stored on the Internet in the form of plaintext,most of which are outsourced to third-party servers,such as online platforms and cloud.However,there are some sensitive words and contents in the official documents.The plaintext data is stored in the untrusted network platform and cloud,which has considerable security risks.Therefore,it is necessary to encrypt the plaintext at the front end,then transmit the ciphertext and store it in the cloud.This process makes the effective use of data,that is,how users retrieve the content they need from the encrypted data,and how to sort the relevance of the retrieved files becomes a very challenging problem.Unstructured data includes land GIS data,office documents in all formats,text,pictures,photos,XML,HTML,various reports,images,audio,video information and other types.However,the files transferred by the authorities and enterprises are mainly stored in the form of unstructured data,such as text files.At present,the research on the ciphertext retrieval technology of unstructured data mainly focuses on the keyword based retrieval of ciphertext data.The retrieval schemes are based on both symmetric and asymmetric encryption systems.The retrieval ideas are divided into sequential scanning method and index retrieval method.However,most of the ciphertext retrieval schemes do not support the ranking of retrieval results by relevance.Although some scholars have proposed a scheme to support the ranking of retrieval results in recent years,the sorting algorithm used in these schemes is too simple,and the quantification of the correlation degree between retrieval statements and result documents is not reasonable.This paper gives the research results in the field of ciphertext at home and abroad,summarizes and explains the classic ciphertext retrieval technology,and based on this,gives the frame structure of ciphertext retrieval in cloud computing environment.This paper is committed to developing as a sub project of "service intelligent office system platform in private cloud environment" led by founder international software(Beijing)Co.,Ltd.It is mainly based on E-government cloud network environment,it focuses on the ciphertext sorting retrieval scheme of text documents and picture text descriptions in the process of document transfer of authorities and enterprises,and improves and optimizes the index construction algorithm,keyword trap algorithm and sorting algorithm,and fully considers the factors such as index keyword weight and query keyword weight in the data,and finally sorts and returns the ciphertext files to the authorized users according to the relevance of query keyword and ciphertext files.In this process,it will not disclose any information about the content of the document and the keywords to be retrieved.The index also hides the cloud service providers to achieve the goal of higher retrieval efficiency,higher security,lower communication cost,and ensure the security of user data while realizing the efficient search of ciphertext data.To meet the transfer and storage of core documents,meetings,business specifications of authorities and enterprises,and realize the needs of future authorities and enterprises for business and security of office platform.Finally,this paper simulate the E-government cloud platform through a private cloud platform,and analyze the security and evaluate the performance of the proposed scheme.From the results,the improved ciphertext sorting and retrieval scheme for unstructured data has high security characteristics,and has advantages over other ciphertext retrieval schemes in communication overhead.