Implementation of Background Knowledge and Properties Induced by Fuzzy Confirmation Measures in Apriori Algorithm
This work contributes to so-called association analysis. Its goal is to search for dependencies (called associations) between attributes in large scale data sets. Recently the authors theoretically studied some properties of fuzzy confirmation measures and possible application of background (resp. expert) knowledge into associations mining process. In this work we implement our recent results into well-known Apriori algorithm. Despite of the fact that the presented algorithm allows us to mine linguistic associations, i.e., associations interpretable in natural language, basic ideas of this algorithm can be easily extended to less specific model of fuzzy sets.
Unable to display preview. Download preview PDF.
- 1.Agrawal, R., Srikant, R.: Fast Algorithms for Mining Association Rules, vol. 1215, pp. 487–499. Citeseer (1994)Google Scholar
- 3.Fu, H.: Cluster analysis and association analysis for the same data, pp. 576–581. University of Cambridge, UK (2008)Google Scholar
- 4.Hájek, P.: The question of a general concept of the guha method. Kybernetika, 505–515 (1968)Google Scholar
- 6.Kupka, J., Tomanová, I.: Some extensions of mining of linguistic associations. Neural Network World 20, 27–44 (2010)Google Scholar
- 7.Kupka, J., Tomanová, I.: Some dependencies among attributes given by fuzzy confirmation measures. In: Proc. of the LFA-EUSFLAT 2011, France, pp. 498–505 (2011)Google Scholar