Title: Post-processing of association rules
Authors: Baesens, Bart
Viaene, Stijn
Vanthienen, Jan
Issue Date: 2000
Publisher: K.U.Leuven
Series Title: DTEW Research Report 0020 pages:1-18
Abstract: In this paper, we situate and motivate the need for a post-processing phase to the association rule mining algorithm when plugged into the knowledge discovery in databases process. Major research effort has already been devoted to optimising the initially proposed mining algorithms. When it comes to effectively extrapolating the most interesting knowledge nuggets from the standard output of these algorithms, one is faced with an extreme challenge, since it is not uncommon to be confronted with a vast amount of association rules after running the algorithms. The sheer multitude of generated rules often clouds the perception of the interpreters. Rightful assessment of the usefulness of the generated output introduces the need to effectively deal with different forms of data redundancy and data being plainly uninteresting. In order to do so, we will give a tentative overview of some of the main post-processing tasks, taking into account the efforts that have already been reported in the literature.
Publication status: published
KU Leuven publication type: IR
Appears in Collections:Research Center for Management Informatics (LIRIS), Leuven

Files in This Item:
File Status SizeFormat
OR_0020.pdf Published 305KbAdobe PDFView/Open


All items in Lirias are protected by copyright, with all rights reserved.