Classification Using Association Rules
Document Type Dissertation
Colin Kane MSc. in Computing (Data Analytics) March 2018
Abstract
This research investigates the use of an unsupervised learning technique, association rules, to make class predictions. The use of association rules to make class predictions is a growing area of focus within data mining research. The research to date has focused predominately on balanced datasets or synthetized imbalanced datasets. There have been concerns raised that the algorithms using association rules to make classifications do not perform well on imbalanced datasets. This research comprehensively evaluates the accuracy of a number of association rule classifiers in predicting home loan sales in an Irish retail banking context. The experiments designed test three associative classifier algorithms CBA, CMAR and SPARCCC against two benchmark algorithms conditional inference trees and random forests on a naturally imbalanced dataset. The experiments implemented and evaluated show that the benchmark tree based algorithms conditional inference trees and random forests outperform the associative classifier models across a range of balanced accuracy measures. This research contributes to the growing body of research in extending association rules to make class predictions. Key words: