Title :
A new method to mine valid association rules
Author :
Luo, Ke ; Wu, Jie
Author_Institution :
Hunan Univ., China
Abstract :
To reduce invalid rules in the mining of association rules, we have analyzed the reasons and presented a relative confidence in the judgment criteria. Based on the value of relative confidence, we classify strong association rules into positive, invalid and negative association rules. We offer an algorithm of mining association rules with new judgment criterion and make tests with Visual FoxPro. The tests indicate that the new method stated in this paper can obviously reduce invalid association rules.
Keywords :
data mining; database management systems; Visual FoxPro; association rules mining; data mining; invalid association rules; judgment criteria; negative association rules; positive association rules; relative confidence; Association rules; Dairy products; Data mining; Distributed databases; Electronic mail; Itemsets; Machine learning; Marketing and sales; Testing; Transaction databases;
Conference_Titel :
Machine Learning and Cybernetics, 2003 International Conference on
Print_ISBN :
0-7803-8131-9
DOI :
10.1109/ICMLC.2003.1264448
