Support driven opportunistic aggregation for generalized itemset extraction

Abstract

Association rule extraction is a widely used exploratory technique which has been exploited in different contexts (e.g., biological data, medical images). However, association rule extraction, driven by support and confidence constraints, entails (i) generating a huge number of rules which are difficult to analyze, or (ii) pruning rare itemsets, even if their hidden knowledge might be relevant. To address the above issues, this paper presents a novel frequent itemset mining algorithm, called GENIO (GENeralized Itemset DiscOverer), to analyze correlation among data by means of generalized itemsets, which provide a powerful tool to efficiently extract hidden knowledge, discarded by previous approaches. The proposed technique exploits a (user provided) taxonomy to drive the pruning phase of the extraction process. Instead of extracting itemsets for all levels of the taxonomy and post-pruning them, the GenIO algorithm performs a support driven opportunistic aggregation of itemsets. Generalized itemsets are extracted only if itemsets at a lower level in the taxonomy are below the support threshold. Experiments performed in the network traffic domain show the efficiency and the effectiveness of the proposed algorithm

Twitter data analysis by means of Strong Flipping Generalized Itemsets

“…To avoid generating all the possible candidates in the taxonomy the authors in [3,41] propose to push (analyst-provided) constraints into the mining process. Many algorithm optimizations have also been proposed [4,13,21,29]. For example, the approach presented in [21] proposes an optimization strategy based on a top-down hierarchy traversal.…”

Personalized tag recommendation based on generalized rules

“…Candidate frequent itemsets are generated by exhaustively evaluating the generalization hierarchies. To reduce the complexity and improve the efficiency of the mining process, several optimization strategies and more efficient algorithms have been proposed [Mennis and Liu 2005;Pramudiono and Kitsuregawa 2004;Srikant et al 1997;Srikant and Agrawal 1995;Han and Fu 2002;Sriphaew and Theeramunkong 2002;Baralis et al 2010]. This article discovers and exploits generalized rules in personalized tag recommendation by adopting an apriori based strategy [Srikant and Agrawal 1995] that integrates, as an itemset mining step, the approach recently proposed in Baralis et al [2010].…”

“…To this aim, two distinct rule sets are generated: (i) a user-specific rule set, which includes all strong generalized rules extracted from the past annotations made by user to which the recommendation is targeted, (ii) a collective rule set, which includes all strong generalized rules mined from the past annotations made by other users. To accomplish the generalized itemset mining task efficiently and effectively, we exploit our implementation of a recently proposed mining algorithm, namely the GENIO algorithm [Baralis et al 2010]. A brief description of the adopted algorithm is given in Section 3.3.1.…”

“…The GENIO Algorithm. GENIO [Baralis et al 2010] is a generalized itemset mining algorithm that addresses the discovery of a smart subset of all the possible frequent (generalized) itemsets. Given a source dataset, a set of generalization hierarchies ρ, and a minimum support threshold minsup, it discovers all frequent not generalized itemsets and all frequent generalized itemsets having at least an infrequent descendant, that is, a descendant that does not satisfy minsup.…”

Misleading Generalized Itemset Mining in the Cloud

“…This technique has already been applied to data coming from different application domains (e.g., market basket analysis [7], network traffic data analysis [8], genetic data mining [9]). Generalized itemset mining entails discovering correlations among data at different abstraction levels.…”