Bayesian classifiers are an important tool in machine learning. The Naive Bayes classifier in particular has shown itself to be a robust, computationally efficient and transparent approximation. However, due to its strong assumption of feature independence it is often too easily discarded. Understanding under what circumstances feature correlations are likely to occur and how to diagnose them is an important first step in improving the Naive Bayes approximation. Furthermore, combining features can lead to enhanced approximations. In this paper we show the benefits of a Generalised Bayes Approximation that accounts for feature correlations while, at the same time, showing that an important area where feature correlations are ubiquitous is in time series that represent aspects of human behaviour. Using data that represent historical patterns of exercise we show how more predictive and more insightful models for obesity can be constructed using a Generalised Bayes approximation that combines historical features, thereby capturing the idea of human habits. In particular, by analysing feature combinations we show that abandoning good past exercise habits is more correlated with obesity than never having had them in the first place.
[1]
John H. Holland,et al.
Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence
,
1992
.
[2]
Igor Kononenko,et al.
Semi-Naive Bayesian Classifier
,
1991,
EWSL.
[3]
Pedro M. Domingos,et al.
Beyond Independence: Conditions for the Optimality of the Simple Bayesian Classifier
,
1996,
ICML.
[4]
Claude Bouchard,et al.
Trends over 5 Decades in U.S. Occupation-Related Physical Activity and Their Associations with Obesity
,
2011,
PloS one.
[5]
J. Lakerveld,et al.
Successful behavior change in obesity interventions in adults: a systematic review of self-regulation mediators
,
2015,
BMC Medicine.
[6]
John H. Holland,et al.
Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence
,
1992
.
[7]
Christopher R. Stephens,et al.
When is the Naive Bayes approximation not so naive?
,
2018,
Machine Learning.