Vol-3 Issue-3 2017 IJARIIE-ISSN(O)-2395-4396 5367 www.ijariie.com 1874 Implementation of Fuzzy Keyword Search Using Encryption & Decryption over Cloud Computing Prof. S.S. Vanjire, Professor at Dept. of CE, Sinhgad Academy of Engineering, Kondhwa, Pune-48 Swapnil Bhapkar 1 , Amardeep Bikkad 2 , Gaurav Bobade 3, Dhiraj Bandle 4 1,2,3,4,- Students, Department of CE, Sinhgad Academy of Engineering, Kondhwa, Pune, Savitribai Phule Pune University, Pune, India ABSTRACT: With the increased rate of growth and adaption of cloud computing, daily, more and more sensitive information is being centralized onto the cloud. For the protection of valuable information, the data must be encrypted before externalization. The existing search techniques allow the user to search over encrypted data using keywords but these techniques account for only exact search. There is no tolerance for typos and format inconsistencies which are normal user behaviour. This makes effective data storage and utilization a very challenging task, supplying user searching very frustrating and inefficient. In this paper, we focus on secure storage using Advanced Encryption Standard (AES) and information retrieval by performing fuzzy keyword search on data which is encrypted while uploading on the cloud. We are proposing the implementation of an advanced fuzzy keyword search mechanism called the Wildcard based techniques which returns the matching files when users searching inputs precisely match the predefined keywords or the closest possible matching files based on similarity keyword semantics when exact match fails. In the proposed solution, we utilize edit distance to quantify keywords similarity and develop an efficient techniques for constructing fuzzy keyword sets, which focus on reducing the storage and representation aloft . KEYWORDS: Fuzzy search, encryption on cloud, cloud computing, AES, wildcard. I. INTRODUCTION Cloud computing is an essential platform for sharing resources. This sharing of resources is based on three models: Infrastructure-as-a-service (IaaS), Platform-as-a-service (PaaS), Software-as-a-service (SaaS).These services are customized as per user demand. Cloud computing more preferably referred as "the cloud”, also focuses on maximizing the effectiveness of shared resources. Usually cloud resources are not only shared by multiple users but are also reallocated as per the user demand. The storage of data on cloud reduces the burden of storage and maintenance of data on the user. Taking into account this tremendous growth of sensitive information on cloud, cloud security is of vital importance for enterprises. The fact that the information owners and vendor provisioned cloud servers are not a part of the same trusted domain may put the outsourced data at risk. This growth of cloud servers are not a part of the same trusted domain may put the outsourced data at risk. This growth of cloud service users has unfortunately been accompanied with a growth in malicious activity in cloud. More and more vulnerabilities are being investigated nearly every day. Millions of users are subscribing to the cloud based services, therefore safety and security of these services are of utmost importance. The future of cloud, even more in expanding range of applications, involves a much deeper degree authentication and privacy. Proposed here is a simple data protection model where data is encrypted using Advanced Encryption Standard (AES) before it is launched into the cloud, thus ensuring data confidentiality and security. To ensure security during information retrieval, we are employing a searchable encryption mechanism.in a standard searchable encryption scheme, an index is created for every keyword of interest and it is associated with the files that contain the keyword. The trapdoors of the keyword are integrated with the index information, thus effective keyword search is realized without compromising file content. In an age of intelligent search systems, the standard searchable encryption scheme supporting an exact keyword match is inconsistent with casual user search behaviours. Normal user search queries will have typos and representation irregularities which may not match the pre-set keyword strings. A user searching for "APPLE" can accidentally type "APLE" and another person may query for “P.O.BOX" because he is ignorant about the stored keyword. Thus, we shift our focus on enabling effective privacy-preserving fuzzy keyword search for information stored in cloud environments. Fuzzy keywords search augments system usability by returning the matching files when users