Articles in Press

Accepted articles here

Latest Issues

Volume 18 Issue 2
Volume 18 Issue 1

Volume 17 Issue 3
Volume 17 Issue 2
Volume 17 Issue 1

Volume 16 Issue 3
Volume 16 Issue 2
Volume 16 Issue 1

Volume 15 Issue 3
Volume 15 Issue 2
Volume 15 Issue 1

Volume 14 Issue 3
Volume 14 Issue 2
Volume 14 Issue 1

Volume 13 Issue 3
Volume 13 Issue 2
Volume 13 Issue 1

Volume 12 Issue 3
Volume 12 Issue 2
Volume 12 Issue 1

Volume 11 Issue 3
Volume 11 Issue 2
Volume 11 Issue 1

Volume 10 Issue 3
Volume 10 Issue 2
Volume 10 Issue 1

Volume 9 Issue 3
Volume 9 Issue 2
Volume 9 Issue 1

Volume 8 Issue 3
Volume 8 Issue 2
Volume 8 Issue 1

Volume 7 Issue 3
Volume 7 Issue 2
Volume 7 Issue 1

Volume 6 Issue 3
Volume 6 Issue 2
Volume 6 Issue 1

Volume 5 Issue 3
Volume 5 Issue 2
Volume 5 Issue 1

Volume 4 Issue 3
Volume 4 Issue 2
Volume 4 Issue 1

Volume 3 Issue 3
Volume 3 Issue 2
Volume 3 Issue 1

Volume 2 Issue 3
Volume 2 Issue 2
Volume 2 Issue 1

Volume 1 Issue 3
Volume 1 Issue 2
Volume 1 Issue 1

Volume 10 Issue 3

A Multi-dimensional Privacy-aware Evaluation Function in Automatic Feature Selection

Yasser Jafer^(a),(*), Stan Matwin^(b),(c), Marina Sokolova^(a),(b),(d)

Transactions on Data Privacy 10:3 (2017) 145 - 174

Abstract, PDF

(a) School of Electrical Engineering and Computer Science, University of Ottawa, Canada.

(b) Institute for Big Data Analytics, Dalhousie University, Canada.

(c) Institute of Computer Science, Polish Academy of Sciences, Poland.

(d) Faculty of Medicine, University of Ottawa, Canada.

e-mail:yjafer @uottawa.ca; stan @cs.dal.ca; sokolova @uottawa.ca

Abstract

Feature selection is based on the notion that redundant and/or irrelevant variables bring no additional information about the data classes and can be considered noise for the predictor. As a result, the total feature set of a dataset could be minimized to only a few features containing maximum discrimination information about the class. Classification accuracy is used as the evaluation measure in guiding the feature selection process. At the same time, such measure does not take into account the privacy of the resulting dataset. In this work, we introduce E(S) a multi-dimensional privacy-aware evaluation function in automatic feature selection that enables the DH to select and eventually release the best subset according to its desired efficacy (e.g., accuracy), privacy, and dimensionality of the resulting dataset.

^* Corresponding author.