Iterative Matrix Factorization Techniques for High-Accuracy Privacy Protection on Non-negative-valued Datasets

1 Abstract— Powerful modern access to huge amounts of various data having high or low level of privacy brings out a concurrent increasing demand for preserving data privacy. The challenge is how to protect attribute values without jeopardizing the similarity between data objects under analysis. In this paper, we further our previous work on applying matrix decomposition techniques to protect privacy and to present a novel algebraic technique based on iterative methods for non-negative-valued data distortion. As an unsupervised learning method for uncovering latent features in high-dimensional data, a low rank nonnegative matrix factorization (NMF) is used to preserve natural data non-negativity and avoid subtractive basis vector and encoding interactions present in techniques such as principal component analysis. This is the first in privacy preserving data mining to combine non-negative matrix decomposition with data distortion processing. Two iterative methods to solve bound-constrained optimization problem in NMF are compared with experiments on Wisconsin Breast Cancer Dataset. The overall performance of NMF on distortion level and data utility is compared to our previously proposed SVD-based distortion strategies and other existing popular data perturbation methods. Data utility is examined by cross validation of a binary classification using support vector machine. Our experimental results indicate that, in comparison with standard data distortion techniques, the proposed NMF-based method is very efficient in balancing data distortion and data utility, and it affords a feasible solution with a good promise on high-accuracy privacy preserving data mining.