I-prune: Item selection for associative classification

Abstract

Associative classification is characterized by accurate models and high model generation time. Most time is spent in extracting and postprocessing a large set of irrelevant rules, which are eventually pruned. We propose I-prune, an item-pruning approach that selects uninteresting items by means of an interestingness measure and prunes them as soon as they are detected. Thus, the number of extracted rules is reduced and model generation time decreases correspondingly. A wide set of experiments on real and synthetic data sets has been performed to evaluate I-prune and select the appropriate interestingness measure. The experimental results show that I-prune allows a significant reduction in model generation time, while increasing (or at worst preserving) model accuracy. Experimental evaluation also points to the chi-square measure as the most effective interestingness measure for item pruning. C 2012 Wiley Periodicals, Inc.

An overview of recent distributed algorithms for learning fuzzy models in Big Data classification

“…As discussed by Baralis et al in [60] and by Abdelhamid et al in [61], Associative Classifiers (ACs) integrate a frequent pattern mining algorithm and a rule-based classifier into a single system. Specifically, first, frequent patterns are extracted from the dataset using an appropriate mining algorithm.…”

Collaborative Filtering Recommendation Algorithm for User Interest and Relationship Based on Score Matrix

“…With the expansion of the scale of e-commerce, the number of users and projects of the system has increased rapidly, but the existing classification information of the project has not changed. Each project has its own attributes that allow users to quickly search for what they need [4]. In general, the item -attribute information of the commodity in the e-commerce system is shown in Figure І.…”

Associative Classification Approaches: Review and Comparison

“…In AC mining, the training phase is about searching for hidden knowledge primarily using association rule algorithms and then a classi¯er is constructed after sorting the knowledge and pruning useless and redundant ones. Many research studies including (Yin and Han, 2003;Thabtah et al, 2005;Li et al, 2008;Ye et al, 2008;Niu et al, 2009;Thabtah et al, 2010;Baralis and Garza, 2012;Abdelhamid et al, 2012a;Zhu et al, 2012;Jabbar et al, 2013;Taiwiah and Sheng, 2013) revealed that AC methods usually extract better classi¯ers with reference to error rate than other classi¯cation data mining approaches like decision tree (Quinlan, 1993) and rule induction (Jensen and Cohen, 2000).…”