A Robustness Measure of Association Rules

We propose a formal definition of the robustness of association rules for interestingness measures. It is a central concept in the evaluation of the rules and has only been studied unsatisfactorily up to now. It is crucial because a good rule (according to a given quality measure) might turn out as a very fragile rule with respect to small variations in the data. The robustness measure that we propose here is based on a model we proposed in a previous work. It depends on the selected quality measure, the value taken by the rule and the minimal acceptance threshold chosen by the user. We present a few properties of this robustness, detail its use in practice and show the outcomes of various experiments. Furthermore, we compare our results to classical tools of statistical analysis of association rules. All in all, we present a new perspective on the evaluation of association rules.

[1]  Philippe Lenca,et al.  A Probabilistic Framework Towards the Parameterization of Association Rule Interestingness Measures , 2007 .

[2]  Tomasz Imielinski,et al.  Mining association rules between sets of items in large databases , 1993, SIGMOD Conference.

[3]  S. Dudoit,et al.  Multiple Testing Procedures with Applications to Genomics , 2007 .

[4]  Bernard Monjardet.  MATHÉMATIQUES ET SCIENCES HUMAINES , 1977 .

[5]  Ramakrishnan Srikant,et al.  Fast Algorithms for Mining Association Rules in Large Databases , 1994, VLDB.

[6]  Philippe Lenca,et al.  A Study of the Robustness of Association Rules , 2007, DMIN.

[7]  Olivier Teytaud,et al.  Association Rule Interestingness: Measure and Statistical Validation , 2007, Quality Measures in Data Mining.

[8]  Einoshin Suzuki,et al.  Pitfalls for Categorizations of Objective Interestingness Measures for Rule Discovery , 2008, Statistical Implicative Analysis.

[9]  Howard J. Hamilton,et al.  Interestingness measures for data mining: A survey , 2006, CSUR.

[10]  Philippe Lenca,et al.  On Optimal Rule Mining: A Framework and a Necessary and Sufficient Condition of Antimonotonicity , 2009, PAKDD.

[11]  P. Lenca,et al.  On the robustness of association rules , 2006, 2006 IEEE Conference on Cybernetics and Intelligent Systems.

[12]  Régis Gras,et al.  Élaboration et évaluation d'un indice d'implication pour des données binaires. I , 1981 .

[13]  Y. Benjamini,et al.  A step-down multiple hypotheses testing procedure that controls the false discovery rate under independence , 1999 .

[14]  Philippe Lenca,et al.  Modeling of the counter-examples and association rules interestingness measures behavior , 2006, DMIN.

[15]  Bruno Crémilleux,et al.  A Unified View of Objective Interestingness Measures , 2007, MLDM.

[16]  Howard J. Hamilton,et al.  Choosing the Right Lens: Finding What is Interesting in Data Mining , 2007, Quality Measures in Data Mining.

[17]  Christian Borgelt,et al.  Induction of Association Rules: Apriori Implementation , 2002, COMPSTAT.

[18]  Patrick Meyer,et al.  On selecting interestingness measures for association rules: User oriented description and multiple criteria decision aid , 2008, Eur. J. Oper. Res..

[19]  Martine Cadot.  A Simulation Technique for extracting Robust Association Rules , 2005 .

[20]  Yves Kodratoff,et al.  Evaluation de la résistance au bruit de quelques mesures d'extraction de règles d'association , 2002, EGC.

[21]  Philippe Lenca,et al.  A Clustering of Interestingness Measures , 2004, Discovery Science.