Equivalence class transformation based mining of frequent itemsets from uncertain data
Carson Kai-Sang Leung, Lijing Sun · 2011
Numerous frequent itemset mining algorithms have been proposed over the past two decades. Most of them mine traditional databases of precise data. However, there are many real-life applications for which data are uncertain. This leads to the mining of uncertain data. In this paper, we propose an equivalence class transformation based algorithm---called UV-Eclat---which transforms probabilistic databases of uncertain data from their usual horizontal format into a vertical format, from which frequent itemsets are mined.