An Improved Rotation-Based Privacy Preserving Classification in Web Mining Using Naïve Bayes Classifier

Subramanian SANGEETHA MARIAMMAL, Ashok KAVITHAMANI

Abstract

Recently, the privacy and security of big data has become an important challenge, which requires the privacy preserving data mining techniques to maintain the trade-off between the data utility and privacy. Web mining is the application of data mining techniques for mining the web data. Privacy issues on the web are based on the fact that most users want to maintain a strict anonymity on web applications and activities. Various conventional techniques are used for privacy preservation like condensation, randomization and tree structure etc. The limitations of the existing approaches are maintaining a balance between the data utility and privacy and the scalability problem. The data stream can be secured and classified by privacy preserving techniques such as perturbation, cryptography and machine learning techniques, etc. Here the machine learning technique Naïve Bayes is employed to classify the perturbed data streams and provides an efficient accuracy. In this proposed research work, the UCI web mining datasets are collected, clustering is done with the help of Fuzzy C-Means (FCM), Flip and Rotation Perturbation (FRP) technique is applied to perform data renovation and classification on perturbed data is done using Naïve Bayes classifier. The classifier is improved by using a holdout approach on data separation in training and testing phases. The classification accuracy, computation time, and error rate of the classifier are measured and they are compared with the ones of the existing method. The comparison shows the achieved result of the proposed method. This proposed system is done with the help of MATLAB 2018a.

Keywords

Web mining, Data perturbation, Flip and Rotation Perturbation, Naïve Bayes classification, Fuzzy C-Means clustering.

View full article