ICCCI'11 Proceedings of the Third international conference on Computational collective intelligence: technologies and applications - Volume Part I
| Hi-index | 0.00 |
Searching for new rules and new knowledge in problem areas,where very little or almostnone previous knowledge is present,can be a very long and emanding process.In ourresearch we addressed the problem of fin ing new knowledge in the form of rules in thediabetes database using a combination of ecision trees and association rules.The firstquestion we wanted to answer was,if there are significant ifferences in sets of rules bothapproaches produce,and how rules,produced by decision trees behave,after being a subject of filtering and reduction,normally used in association rule approaches.In order to accomplishthat,we had to make some modifications to both the decision tree approach and associationrule approach.From the first results we can conclude,that the sets of rules,built by decisiontrees are much smaller than the sets created by association rules.We coul also establish,thatfiltering and reduction did not effect the rules derived from decision trees in the same scale asassociation rules.