A robustness measure of association rules
作者： Yannick Le Bras , Patrick Meyer , Philippe Lenca , Stéphane Lallich
DOI: 10.1007/978-3-642-15883-4_15
关键词:
摘要: We propose a formal definition of the robustness association rules for interestingness measures. It is central concept in evaluation and has only been studied unsatisfactorily up to now. crucial because good rule (according given quality measure) might turn out as very fragile with respect small variations data. The measure that we here based on model proposed previous work. depends selected measure, value taken by minimal acceptance threshold chosen user. present few properties this robustness, detail its use practice show outcomes various experiments. Furthermore, compare our results classical tools statistical analysis rules. All all, new perspective
springer.com
archives-ouvertes.fr
sci-hub.st 参考文章(24)
Philippe Lenca, Stéphane Lallich, Benoît Vaillant, Modeling of the counter-examples and association rules interestingness measures behavior DMIN. pp. 132- 137 ,(2006)
Philippe Lenca, Stéphane Lallich, Benoît Vaillant, Jérôme Azé, A study of the robustness of association rules international conference on data mining. pp. 163- 169 ,(2007)
Yves Kodratoff, Jérôme Azé, Evaluation de la résistance au bruit de quelques mesures d'extraction de règles d'association. EGC. pp. 143- 154 ,(2002)
Ramakrishnan Srikant, Rakesh Agrawal, Fast Algorithms for Mining Association Rules in Large Databases very large data bases. pp. 487- 499 ,(1994)
Christian Borgelt, Rudolf Kruse, Induction of Association Rules: Apriori Implementation COMPSTAT. pp. 395- 400 ,(2002) , 10.1007/978-3-642-57489-4_59
Yannick Le Bras, Philippe Lenca, Stéphane Lallich, On Optimal Rule Mining: A Framework and a Necessary and Sufficient Condition of Antimonotonicity Advances in Knowledge Discovery and Data Mining. pp. 705- 712 ,(2009) , 10.1007/978-3-642-01307-2_71
Stephane Lallich, Olivier Teytaud, Elie Prudhomme, Association Rule Interestingness: Measure and Statistical Validation Quality Measures in Data Mining. pp. 251- 275 ,(2007) , 10.1007/978-3-540-44918-8_11
Einoshin Suzuki, Pitfalls for categorizations of objective interestingness measures for rule discovery Statistical Implicative Analysis. pp. 383- 395 ,(2008) , 10.1007/978-3-540-78983-3_17
Benoît Vaillant, Philippe Lenca, Stéphane Lallich, A Clustering of Interestingness Measures discovery science. pp. 290- 297 ,(2004) , 10.1007/978-3-540-30214-8_23
Liqiang Geng, Howard J. Hamilton, Choosing the Right Lens: Finding What is Interesting in Data Mining Quality Measures in Data Mining. pp. 3- 24 ,(2007) , 10.1007/978-3-540-44918-8_1