Databases are used among various areas for storage of large data items. Viewing data from this large dataset became a difficult task. Hence, data mining was introduced to retrieve and view the necessary data. Data mining is a process of extracting the large databases, these databases are mostly scattered among various websites where security is required. These databases are divided into various categories such as horizontally distributed databases, vertically distributed databases and hybrid distributed databases. When this data size increases, there exists data leakage, less security, etc. The distributed databases are shared among various servers where insecurity occurs. To secure databases when sharing data in a database various methodologies were being used. Earlier there exists a trusted third party who performs various cryptographic primitives and the trusted third party acts as an intermediate between the sites in which the probability of data leakage increases. Hence, this paper proposes a new protocol to reduce the data leakage and to improve the security of the horizontally distributed databases. The outcome of the proposed architecture provides methods where no trusted third party is required, the sites themselves communicate each other for a secure mining as well as enhanced security. This proposed architecture covers almost all the disadvantages occurred in the previous algorithms and has been implemented in the synthetic employment office database.

No abstract

No abstract

No abstract

This paper is a survey of the big data idea, its measurements, its design correlation between the prior idea and the most recent, the capacity conceivable i.e. the databases and starting point of enormous information. The social database, having unbending pattern, has been winning since quite a while yet it is hard to store the unstructured information in social database. The unstructured information has principally message nature or is as logs. Here comes the idea of No SQL databases. Enormous Data is little information with huge information estimate.