Quantifying and Visualizing Attribute Interactions

TL;DR

This paper discusses the quantification and visualization of attribute interactions in data using interaction information, demonstrating how these insights can improve data understanding and predictive modeling.

Contribution

It introduces the use of McGill's interaction information for visualizing attribute interactions and illustrates its application across various data domains.

Findings

Interaction information effectively visualizes key attribute interactions.

Visualization reveals data structure, redundancies, and regularities.

Methods assist in feature construction and model improvement.

Abstract

Interactions are patterns between several attributes in data that cannot be inferred from any subset of these attributes. While mutual information is a well-established approach to evaluating the interactions between two attributes, we surveyed its generalizations as to quantify interactions between several attributes. We have chosen McGill's interaction information, which has been independently rediscovered a number of times under various names in various disciplines, because of its many intuitively appealing properties. We apply interaction information to visually present the most important interactions of the data. Visualization of interactions has provided insight into the structure of data on a number of domains, identifying redundant attributes and opportunities for constructing new features, discovering unexpected regularities in data, and have helped during construction of predictive models; we illustrate the methods on numerous examples. A machine learning method that disregards interactions may get caught in two traps: myopia is caused by learning algorithms assuming independence in spite of interactions, whereas fragmentation arises from assuming an interaction in spite of independence.