Attribute-Based Document Retrieval with Modified ARF Tree in Cloud Computing

Safe storage and retrieval of document is one of the mostsought after research directions in cloud computing. Inspite of many searchable encryption schemes being put forward, seldom of them supports skillful and productive retrieval over the documents which are encrypted based on their attributes. In this project, a hierarchical attribute-based encryption scheme is primarily designed for document collection. A lot of records can be scrambled together in the event that they are based on a coordinated access structure. On scrutinizing with the Cipher text-Policy Attribute-Based Encryption (CP-ABE) schemes, both the cipher text storage space and time costs of encryption/decryption are conserved. At that point, a record structure named Attribute-based Retrieval Features (ARF) tree produced for the report gathering dependent on the TF-IDF show and the archives properties. A profundity first scan calculation for the ARF tree is intended to improve the inquiry proficiency which can be additionally improved by parallel processing. With the exception of the report accumulations, our plan can be used for different datasets by changing the ARF tree imperceptibly. A meticulous analysis and an array of experiments are performed to emblematize the security and efficacy of the proposed scheme.