Association Rule Mining And Data Sanitization Technique Using Hash Base Apriori
The private information is compassionate under different circumstances and needs to be sanitized before it is released to resolve privacy concerns. With limited time, data mining techniques can obtain vast quantities of information. Most sensitive material corresponds to an individual or an organization, the information obtained by the sophisticated machine learning algorithm could disclose. Data belonging to an entity or an agency can have different degrees of sensitivity. Such data is only made accessible to approved users or entities. Therefore, confirming the security of complex data by access constraints is not a complete operation. It can affect the usefulness of the product of data mining, and the user can re-identify sensitive data. To find a mechanism for the security of confidential information by introducing tools and techniques for data mining that could be applied to databases, despite it is compromising the reliability of the outcome of data mining. In this article, we proposed data sanitization technique using frequent itemset classification approach with modified apriori algorithm. The difficulty is on maintaining actionable intelligence for necessary arrangements, but at the corresponding period not suffering the numerous exhibition of corporation rule mining. Thorough analysis of various sequential pattern algorithms for achieving the privacy of extensive data using data sanitization technique. Our investigations illustrate that our algorithm is efficient, scalable, and performs meaningful correction over the other methods conferred in the existing systems.