Utility-based association rule mining: A marketing solution for cross-selling

Cited 39 time in webofscience Cited 0 time in scopus
  • Hit : 1543
  • Download : 0
Recently, a utility-based mining approach has emerged as an alternative mechanism to frequency-based mining in an attempt to reflect not only the statistical correlation but also the semantic significance (e.g., price and quantity) of items. However, existing mining trajectories utilizing high-utility itemsets may not offer firms sufficient business insights unless they can precisely assess the value of association rules, which may vary substantially depending on many business parameters included in the assessment. In this study, we propose a utility-based association-rule mining method that valuates association rules by measuring their specific business benefits accruing to firms. Based on previous studies, three key elements (opportunity, effectiveness, and probability) are identified to define and operationalize a users' preference as a utility function. To apply the utility-based mechanism to the processing of large transaction databases, we constructed functional algorithms, with heightened attention paid to their pruning strategies, and evaluated them based on real-world databases. Experimental results show that the proposed approach can provide users with greater business benefits than the high-utility itemset mining approach, suggesting several important strategic implications for both research and practice. (C) 2012 Elsevier Ltd. All rights reserved.
Publisher
PERGAMON-ELSEVIER SCIENCE LTD
Issue Date
2013-06
Language
English
Article Type
Article
Citation

EXPERT SYSTEMS WITH APPLICATIONS, v.40, no.7, pp.2715 - 2725

ISSN
0957-4174
DOI
10.1016/j.eswa.2012.11.021
URI
http://hdl.handle.net/10203/173480
Appears in Collection
MT-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 39 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0