Selecting the right interestingness measure for association patterns Export

@inproceedings{citeulike:5230250, abstract = {Many techniques for association rule mining and feature selection require a suitable metric to capture the dependencies among variables in a data set. For example, metrics such as support, confidence, lift, correlation, and collective strength are often used to determine the interestingness of association patterns. However, many such measures provide conflicting information about the interestingness of a pattern, and the best metric to use for a given application domain is rarely known. In this paper, we present an overview of various measures proposed in the statistics, machine learning and data mining literature. We describe several key properties one should examine in order to select the right measure for a given application domain. A comparative study of these properties is made using twenty one of the existing measures. We show that each measure has different properties which make them useful for some application domains, but not for others. We also present two scenarios in which most of the existing measures agree with each other, namely, support-based pruning and table standardization. Finally, we present an algorithm to select a small set of tables such that an expert can select a desirable measure by looking at just this small set of tables.}, address = {New York, NY, USA}, author = {Tan, Pang N. and Kumar, Vipin and Srivastava, Jaideep}, booktitle = {KDD '02: Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining}, citeulike-article-id = {5230250}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=775047.775053}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/775047.775053}, doi = {10.1145/775047.775053}, isbn = {1-58113-567-X}, keywords = {association-rules, interestingness-measures, validation}, location = {Edmonton, Alberta, Canada}, pages = {32--41}, posted-at = {2009-11-04 19:41:15}, priority = {0}, publisher = {ACM}, title = {Selecting the right interestingness measure for association patterns}, url = {http://dx.doi.org/10.1145/775047.775053}, year = {2002} }

TY - CONF ID - citeulike:5230250 L3 - citeulike-article-id:5230250 N2 - Many techniques for association rule mining and feature selection require a suitable metric to capture the dependencies among variables in a data set. For example, metrics such as support, confidence, lift, correlation, and collective strength are often used to determine the interestingness of association patterns. However, many such measures provide conflicting information about the interestingness of a pattern, and the best metric to use for a given application domain is rarely known. In this paper, we present an overview of various measures proposed in the statistics, machine learning and data mining literature. We describe several key properties one should examine in order to select the right measure for a given application domain. A comparative study of these properties is made using twenty one of the existing measures. We show that each measure has different properties which make them useful for some application domains, but not for others. We also present two scenarios in which most of the existing measures agree with each other, namely, support-based pruning and table standardization. Finally, we present an algorithm to select a small set of tables such that an expert can select a desirable measure by looking at just this small set of tables. CY - Edmonton, Alberta, Canada SN - 1-58113-567-X AD - New York, NY, USA EP - 41 TI - Selecting the right interestingness measure for association patterns PB - ACM T2 - KDD '02: Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining SP - 32 KW - association-rules KW - interestingness-measures KW - validation AU - Tan, Pang AU - Kumar, Vipin AU - Srivastava, Jaideep PY - 2002/// UR - http://dx.doi.org/10.1145/775047.775053 ER -