Ranking Association Rules by Clustering Through Interestingness
The association rules (ARs) post-processing step is challenging, since many patterns are extracted and only a few of them are useful to the user. One of the most traditional approaches to find rules that are of interestingness is the use of objective measures (OMs). Due to their frequent use, many of them exist (over 50). Therefore, when a user decides to apply such strategy he has to decide which one to use. To solve this problem this work proposes a process to cluster ARs based on their interestingness, according to a set of OMs, to obtain an ordered list containing the most relevant patterns. That way, the user does not need to know which OM to use/select nor to handle the output of different OMs lists. Experiments show that the proposed process behaves equal or better than as if the best OM had been used.
KeywordsAssociation rules Post-processing Objective measures Clustering
We wish to thank FAPESP (2015/08059-0) and CAPES for the financial support.
- 2.Bong, K.K., Joest, M., Quix, C., Anwar, T.: Automated interestingness measure selection for exhibition recommender systems. In: Nguyen, N.T., Attachoo, B., Trawiński, B., Somboonviwat, K. (eds.) ACIIDS 2014. LNCS (LNAI), vol. 8397, pp. 221–231. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-05476-6_23CrossRefGoogle Scholar
- 3.de Carvalho, V.O., de Padua, R., Rezende, S.O.: Solving the problem of selecting suitable objective measures by clustering association rules through the measures themselves. In: Freivalds, R.M., Engels, G., Catania, B. (eds.) SOFSEM 2016. LNCS, vol. 9587, pp. 505–517. Springer, Heidelberg (2016). https://doi.org/10.1007/978-3-662-49192-8_41CrossRefzbMATHGoogle Scholar
- 4.de Carvalho, V.O., dos Santos, F.F., Rezende, S.O., de Padua, R.: PAR-COM: a new methodology for post-processing association rules. In: Zhang, R., Zhang, J., Zhang, Z., Filipe, J., Cordeiro, J. (eds.) ICEIS 2011. LNBIP, vol. 102, pp. 66–80. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-29958-2_5CrossRefGoogle Scholar
- 5.Djenouri, Y., Drias, H., Habbas, Z., Chemchem, A.: Organizing association rules with meta-rules using knowledge clustering. In: 11th International Symposium on Programming and Systems, pp. 109–115 (2013)Google Scholar
- 6.Ester, M., Kriegel, H., Sander, J., Xu, X.: A density-based algorithm for discovering clusters in large spatial databases with noise. In: 2nd International Conference on Knowledge Discovery and Data Mining, pp. 226–231 (1996)Google Scholar
- 7.Guillaume, S., Grissa, D., Mephu Nguifo, E.: Categorization of interestingness measures for knowledge extraction. CoRR abs/1206.6741 (2012)Google Scholar
- 8.Huynh, X.H., Guillet, F., Blanchard, J., Kuntz, P., Briand, H., Gras, R.: A graph-based clustering approach to evaluate interestingness measures: a tool and a comparative study. In: Guillet, F.J., Hamilton, H.J. (eds.) Quality Measures in Data Mining. Studies in Computational Intelligence, vol. 43, pp. 25–50. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-44918-8_2CrossRefGoogle Scholar
- 10.Tan, P., Steinbach, M., Kumar, V.: Introduction to Data Mining (2006)Google Scholar