Enhancing Distributed Data Integrity Verification Scheme in Cloud Environment Using Machine Learning Approach

Abstract

DML (Distributed machine learning) is considered one of the most important AI technologies (AI). Nonetheless, data integrity isn't taken into account in the present distributed machine learning configuration. If network intruders falsify the data, however, change or destroy the data. As a result, ensuring data integrity in the DML is critical. To verify that the training data is accurate, it presents a distributed machine-learned ness initiates data integrity checking method. To begin, it advocates for the implementation of a Provable Data Possession (PDP) selection checking method to obtain data integrity confirmation, as a result of DML-DIV technique will be able to resist manipulation as well as fraud assaults. Second, in the TPA verification process, researchers produce a unique integer, called the blinded variable, and use the discrete logarithm problem (DLP) to establish evidence and assure data protection. To invoke the data possessor's public/private set of keys, it uses two-step authentication and identity-based cryptogra phic pivotal generating technology. As a result of which our DML-DIV scheme may be able to tackle the critical escrow issue while also lowering document management expenses. Lastly, a rigorous theoretical analysis is performed, formal theoretical analysis and empirical data confirm the safety and effectiveness of the DML-DN system.