Abstract: The problem of reliability auditing and protected deduplication on reasoning information. Particularly, seeking at achieving both information reliability and deduplication in reasoning, we propose two protected techniques, namely SecCloud and SecCloud⁺. SecCloud presents an review enterprise with a servicing of a MapReduce reasoning which helps clients produce information labels before uploading as well as review the reliability of information having been stored in reasoning. In comparison with past work, the calculations by user in SecCloud is decreased during the data file posting and auditing stages. SecCloud is developed inspired by the fact that customers always want to secure their information before posting and allows reliability review and protected deduplication on encrypted data. So, propose to use active learning genetic programming mechanism a query-dependent record matching method that requires semi supervised data set. Active learning approach is used in AGP in which a committee of multi attribute functions votes for classifying record pairs as duplicates or not. Quality of record can show the results of AGP deduplication while reducing the number of labeled examples was needed.