Mining association rules between sets of items in large databases
SIGMOD '93 Proceedings of the 1993 ACM SIGMOD international conference on Management of data
An efficient graph-based approach to mining association rules for large databases
International Journal of Intelligent Information and Database Systems
Learning task models in ill-defined domain using an hybrid knowledge discovery framework
Knowledge-Based Systems
Measures for comparing association rule sets
ICAISC'10 Proceedings of the 10th international conference on Artificial intelligence and soft computing: Part I
Interestingness measures for association rules based on statistical validity
Knowledge-Based Systems
ShrFP-tree: an efficient tree structure for mining share-frequent patterns
AusDM '08 Proceedings of the 7th Australasian Data Mining Conference - Volume 87
| Hi-index | 0.00 |
Mining association rules are widely studied in data mining society. In this paper, we analyze the measure method of support-confidence framework for mining association rules, from which we find it tends to mine many redundant or unrelated rules besides the interesting ones. In order to ameliorate the criterion, we propose a new method of match as the substitution of confidence. We analyze in detail the property of the proposed measurement. Experimental results show that the generated rules by the improved method reveal high correlation between the antecedent and the consequent when the rules were compared with that produced by the support-confidence framework. Furthermore, the improved method decreases the generation of redundant rules.