Misleading Generalized Itemset discovery

作者：

Highlights：

• A novel approach to discovering valuable knowledge from large datasets is proposed.

• We compare data correlations at different abstraction levels.

• An algorithm to extract misleading high-level data correlations is presented.

• Itemsets representing contrasting data correlations are discovered.

• The effectiveness of the proposed approach has been evaluated on real mobile data.

摘要

•A novel approach to discovering valuable knowledge from large datasets is proposed.•We compare data correlations at different abstraction levels.•An algorithm to extract misleading high-level data correlations is presented.•Itemsets representing contrasting data correlations are discovered.•The effectiveness of the proposed approach has been evaluated on real mobile data.

论文关键词：Generalized itemset mining,Data mining,Taxonomies,Mobile data analysis

论文评审过程：Available online 29 August 2013.

论文官网地址：https://doi.org/10.1016/j.eswa.2013.08.039