Maintaining Privacy for Multi-keyword Search on Encrypted Data:A Survey

Cloud computing is  rising as a promising pattern for information outsourcing and high-quality services. The data owners prefer to outsource their data to the cloud server in order to reduce data management cost and storage facility. Data encryption and compression is used to maintain the security and privacy of documents and also to reduce the cloud storage space. The encrypted documents are stored on cloud server. The similarity between the documents will be hiding in the process of encryption, which will lead to the search accuracy performance degradation. The amount of data stored on cloud server increases per day. It is challenging to design search technique on encrypted documents that can provide the reliable online information retrieval on large volume of encrypted data. We propose a cosine similarity clustering to support more search semantics and fast search within the large volume of data. The proposed cosine similarity approach clusters the documents based on their cosine similarity value. We also propose a method which can maintain and utilize the relationship between documents to increase the speed of search.