Confidential Data Identification Using Data Mining Techniques in Data Leakage Prevention System
Subhashini Peneti, Padmaja Rani B · International Journal of Data Mining & Knowledge Management Process · 2015
Data leakage means sending confidential data to an unauthorized person.Nowadays, identifying confidential data is a big challenge for the organizations.We developed a system by using data mining techniques, which identifies confidential data of an organization.First, we create clusters for the training data set.Next, identify confidential terms and context terms for each cluster.Finally, based on the confidential terms and context terms, the confidentiality level of the detected document calculated in terms of score.If the score of the detected document beyond a predefined threshold, then the document is blocked and marked as a confidential.