Due to the combinatorial nature of the problem, the proposed methodologies span from simple, time and memory efficient heuristics and border-based approaches, to exact hiding algorithms that offer guarantees on the quality of the computed hiding solution at an increased, however, computational complexity cost.

Course Objectives

This course provides an extensive survey on a specific class of privacy preserving data mining methods that belong to the knowledge hiding thread and are collectively known as association rule hiding methods.

Recommended or Required Reading

A.G.GKOULALAS-DIVANIS and V.S. VERYKIOS. Association rule hiding functions, Springer, 2010. ,O. ABUL, M. ATZORI, F. BONCHI, and F. GIANNOTTI. Hiding sequences, In Proceedings of the 23rd International Conference on Data Engineering Workshops (ICDEW), pp.147–156, 2007. ,C. C. AGGARWAL and P. S. YU. Privacy preserving data mining: Models and algorithms, Springer–Verlag, 2008.

Learning Outcomes

1. Understand and compare the approach and algorithms of privacy preserving data mining

2. Make a survey on one of the open problems of the field

3. Model and implement one of the tasks of privacy preserving data mining

4. Document and present the project of the developed

Grading

Midterm 30%

Homework 30%

Final 40%