International Research Journal of Engineering and Technology (IRJET) Volume: 04 Issue: 03 | Mar -2017
e-ISSN: 2395 -0056 p-ISSN: 2395-0072
www.irjet.net
Efficient Similarity Search over Encrypted Data Mohit Kulkarni1, Nikhil Kumar2, Santosh Vaidande3, B.S.Satpute4 1Mohit
Kulkarni: Student, Dept. of Computer Engineering, Dr. D. Y. Patil Institute of Technology, Maharashtra, India. 2Nikhil Kumar: Student, Dept. of Computer Engineering, Dr. D. Y. Patil Institute of Technology, Maharashtra, India. 3Santosh Vaidande: Student, Dept. of Computer Engineering, Dr. D. Y. Patil Institute of Technology, Maharashtra, India. 4B.S. Satpute: Professor, Dept. of Computer Engineering, Dr. D. Y. Patil Institute of Technology, Maharashtra, India. ---------------------------------------------------------------------***--------------------------------------------------------------------the operation which is called Searchable Encryption Abstract – In the present time, due to attractive Scheme. features of cloud computing, the massive amount of data has been stored in the cloud. Though cloud-based Conventionally, almost all schemes are designed for services offer many benefits but privacy and security of exact query matching. Query matching is the process of the sensitive data is a big issue. These issues are resolved finding the user’s desired data from the cloud by storing sensitive data in encrypted form. Encrypted according to the stated feature. But actually, it is more storage protects the data against unauthorized access, logical to perform retrieval according to the similarity but it weakens some basic and important functionality with the stated feature instead of the existence of it. like search operation on the data, i.e. searching the required data by the user on the encrypted data requires Similarity search is a problem that optimizes and finds data to be decrypted first and then search, so this the point in a given set that is closest to a given point. eventually, slows down the process of searching. To It is predefined that efficient methods are required to achieve this many encryption schemes have been do similarity search over the huge amount of encrypted proposed, however, all of the schemes handle exact data. The basic algorithm of our project is the Query matching but not Similarity matching. While user approximate near neighbor search algorithm called uploads the file, features are extracted from each Locality Sensitive Hashing (LSH). LSH is broadly used document. When the user fires a query, trapdoor of that for fast similarity search on data in information query is generated and search is performed by finding retrieval. Our project has two parts: User ‘Uploading the correlation among documents stored on cloud and the data’ and ‘Searching the query’. query keyword, using Locality Sensitive Hashing. Key Words - Locality sensitive hashing, Encrypted data, Similarity search, Cloud computing, Data privacy and Query matching. 1. INTRODUCTION In present data comprehensive environment, cloud computing is common because it removes the burden of enormous data management in a cost effective way. The sensitive data send to untrusted cloud servers will lead to privacy issues. To lessen the worries, sensitive data must be deployed in the encrypted form which avoids illegal access. Though encryption provides protection, it complicates the fundamental search operation. There are many algorithms which support
© 2017, IRJET
|
Impact Factor value: 5.181
|
Fig 1: Complete Process
ISO 9001:2008 Certified Journal
|
Page 2133