Testing the Significance of Attribute Interactions

Aleks Jakulin and Ivan Bratko (2004) Testing the Significance of Attribute Interactions. In: Twenty-first International Conference on Machine Learning (ICML-2004), July, 2004, Banff, Canada.

Preview

PDF Download (294Kb)

Abstract

Attribute interactions are the irreducible dependencies between attributes. Interactions underlie feature relevance and selection, the structure of joint probability and classification models: if and only if the attributes interact, they should be connected. While the issue of 2-way interactions, especially of those between an attribute and the label, has already been addressed, we introduce an operational definition of a generalized $n$-way interaction by highlighting two models: the reductionistic part-to-whole approximation, where the model of the whole is reconstructed from models of the parts, and the holistic reference model, where the whole is modelled directly. An interaction is deemed significant if these two models are significantly different. In this paper, we propose the Kirkwood superposition approximation for constructing part-to-whole approximations. To model data, we do not assume a particular structure of interactions, but instead construct the model by testing for the presence of interactions. The resulting map of significant interactions is a graphical model learned from the data. We confirm that the P-values computed with the assumption of the asymptotic chi^2 distribution closely match those obtained with the bootstrap.

Actions (login required)