Association rule mining is a widely used data mining technique, but can provide an overwhelming number of discovered rules. The objective of this project was to investigate the post-processing of association rules to automatically select the most relevant rules and facilitate feasible human analysis. The post-processing techniques used included pruning uninteresting rules, pruning redundant rules and summarizing rules with Direction Setting Rules. These techniques were implemented in the WPI-WEKA data mining system and then tested for efficiency and effectiveness.

• This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator

• Paredes, Eduardo Javier.

Publisher

• Worcester Polytechnic Institute

Identifier

• 05C019M

Advisor

• Ruiz, Carolina

Year

• 2005

Date created

• 2005-01-01

Resource type

• Major Qualifying Project

Major

• Computer Science

Rights statement

• In Copyright