QROC: A Variation of ROC Space to Analyze Item Set Costs/Benefits in Association Rules

Receiver Operating Characteristics (ROC) graph is a popular way of assessing the performance of classification rules. However, as such graphs are based on class conditional probabilities, they are inappropriate to evaluate the quality of association rules. This follows from the fact that there is no class in association rule mining, and the consequent part of two different association rules might not have any correlation at all. This chapter presents an extension of ROC graphs, named QROC (for Quality ROC), which can be used in association rule context. Furthermore, QROC can be used to help analysts to evaluate the relative interestingness among different association rules in different cost scenarios.

[1]  Peter A. Flach.  The Geometry of ROC Space: Understanding Machine Learning Metrics through ROC Isometrics , 2003, ICML.

[2]  Heikki Mannila,et al.  Association Rule Selection in a Data Mining Environment , 1999, PKDD.

[3]  A. Bradley.  ROC curves and the X2 test , 1996, Pattern Recognit. Lett..

[4]  Helena C. Kraemer,et al.  Assessment of 2 × 2 Associations: Generalization of Signal-Detection Methodology , 1988 .

[5]  Gregory Piatetsky-Shapiro,et al.  Discovery, Analysis, and Presentation of Strong Rules , 1991, Knowledge Discovery in Databases.

[6]  Andrew P. Bradley,et al.  The use of the area under the ROC curve in the evaluation of machine learning algorithms , 1997, Pattern Recognit..

[7]  Fabrice Guillet,et al.  Quality Measures in Data Mining (Studies in Computational Intelligence) , 2007 .

[8]  Tom Fawcett,et al.  An introduction to ROC analysis , 2006, Pattern Recognit. Lett..

[9]  Tom Fawcett.  PRIE: a system for generating rulelists to maximize ROC performance , 2008, Data Mining and Knowledge Discovery.

[10]  Robert C. Holte,et al.  Cost curves: An improved method for visualizing classifier performance , 2006, Machine Learning.

[11]  Tom Fawcett,et al.  Robust Classification for Imprecise Environments , 2000, Machine Learning.

[12]  Geert Wets,et al.  Using association rules for product assortment decisions: a case study , 1999, KDD '99.

[13]  Wynne Hsu,et al.  Pruning and summarizing the discovered associations , 1999, KDD '99.

[14]  Jaideep Srivastava,et al.  Selecting the right interestingness measure for association patterns , 2002, KDD.

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

[16]  Abdelaziz Berrado,et al.  Using metarules to organize and group discovered association rules , 2006, Data Mining and Knowledge Discovery.

[17]  Helena Chmura Kraemer,et al.  Evaluating Medical Tests: Objective and Quantitative Guidelines , 1992 .

[18]  Fabrice Guillet,et al.  Quality Measures in Data Mining , 2009, Studies in Computational Intelligence.

[19]  Ramakrishnan Srikant,et al.  Mining Association Rules with Item Constraints , 1997, KDD.

[20]  Gerd Stumme,et al.  Generating a Condensed Representation for Association Rules , 2005, Journal of Intelligent Information Systems.

[21]  Laks V. S. Lakshmanan,et al.  Exploratory mining via constrained frequent set queries , 1999, SIGMOD '99.

[22]  Ron Kohavi,et al.  The Case against Accuracy Estimation for Comparing Induction Algorithms , 1998, ICML.

[23]  James P. Egan,et al.  Signal detection theory and ROC analysis , 1975 .

[24]  Peter A. Flach,et al.  A Response to Webb and Ting’s On the Application of ROC Analysis to Predict Classification Performance Under Varying Class Distributions , 2005, Machine Learning.

[25]  Peter A. Flach,et al.  ROCCER: An Algorithm for Rule Learning Based on ROC Analysis , 2005, IJCAI.

[26]  Hiroyuki Kawano,et al.  Extended Association Algorithm Based on ROC Analysis for Visual Information Navigator , 2002, Progress in Discovery Science.