Mining association rules between sets of items in large databases
SIGMOD '93 Proceedings of the 1993 ACM SIGMOD international conference on Management of data
Mining the most interesting rules
KDD '99 Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining
Mining frequent patterns without candidate generation
SIGMOD '00 Proceedings of the 2000 ACM SIGMOD international conference on Management of data
Mining Pareto-optimal rules with respect to support and confirmation or support and anti-support
Engineering Applications of Artificial Intelligence
Assessing the Quality of Rules with a New Monotonic Interestingness Measure Z
ICAISC '08 Proceedings of the 9th international conference on Artificial Intelligence and Soft Computing
Analysis of symmetry properties for bayesian confirmation measures
RSKT'12 Proceedings of the 7th international conference on Rough Sets and Knowledge Technology
Hi-index | 0.00 |
Evaluating the interestingness of rules or trees is a challenging problem of knowledge discovery and data mining. In recent studies, the use of two interestingness measures at the same time was prevailing. Mining of Pareto-optimal borders according to support and confidence, or support and anti-support are examples of that approach. Here, we consider induction of "if..., then..." association rules with a fixed conclusion. We investigate ways to limit the set of rules non---dominated wrt support and confidence or support and anti-support, to a subset of truly interesting rules. Analytically, and through experiments, we show that both of the considered sets can be easily reduced by using the valuable semantics of confirmation measures.