Multi Keyword Similarity Search Over Encrypted Text Data on Cloud

Abstract

The tremendous amount of data is being outsourced every day by individuals or enterprises . It is not feasible to manage or to store such a large data locally, due to the limited storage capacities, and the system becomes the single point of failure. the cloud comes into picture to store the data with better flexibility and cost saving. As the data might be confidential or sensitive, the data which user wants to store on the cloud can be private and it should not be leaked, for that purpose searchable encryption is be used, so that even if the file falls in wrong hands it will be safe. At the time of retrieval of data, the multi-keyword search over text data can only handle the exact keywork matching. Multi-keyword similarity search overcomes the problem of not finding any related documents on searching. while encrypting the data before storing it to the cloud will help to preserve the privacy of the files. Finding the similarities between input keyword or similar keyword is done by edit distance metric algorithm. Final design to achieve the user privacy, and to speedup the search task. At cloud side Bloom Inverted List is used to implement searching on index.