Significant Pattern Mining with Covariates (FACS)

Main content

Laetitia Papaxanthos, Felipe Llinares-López, Dean Bodenham and Karsten Borgwardt

Finding significant combinations of features in the presence of categorical covariates

Summary

In this project, we developed the first approach to significant discriminative itemset mining that allows one to correct for a confounding categorical covariate.

We propose the Fast Automatic Conditional Search (FACS) algorithm, a significant discriminative itemset mining method which conditions on categorical covariates and only scales as O(k log k), where k is the number of states of the categorical covariate. Based on the Cochran-Mantel-Haenszel Test, FACS demonstrates superior speed and statistical power on simulated and real-world datasets compared to the state of the art, opening the door to numerous applications in biomedicine.

The following short spotlight video also summarizes the main aspects of the paper:

Code

The code is available here on Github.

Reference

Keyboard navigation between tabs via Alt arrow keys as well as Home and End.

Laetitia Papaxanthos*, Felipe Llinares-López*, Dean Bodenham and Karsten Borgwardt (*=equal contributions)
Finding significant combinations of features in the presence of categorical covariates,

Advances in Neural Information Processing Systems 29 (NIPS 2016), 2271-2279. (Online)

@incollection{Papaxanthos-2016-NIPS,
title = {Finding significant combinations of features in the presence of categorical covariates},
author = {Papaxanthos, Laetitia and Llinares-L\'{o}pez, Felipe and Bodenham, Dean and Borgwardt, Karsten},
booktitle = {Advances in Neural Information Processing Systems 29},
editor = {D. D. Lee and M. Sugiyama and U. V. Luxburg and I. Guyon and R. Garnett},
pages = {2271--2279},
year = {2016},
publisher = {Curran Associates, Inc.},
url = {http://papers.nips.cc/paper/6345-finding-significant-combinations-of-features-in-the-presence-of-categorical-covariates.pdf}
}

Further information and the code can be found on the project page.

Contact for questions regarding usage or reporting bugs.

 
 
Page URL: https://www.bsse.ethz.ch/mlcb/research/machine-learning/facs.html
27.04.2017
© 2017 Eidgenössische Technische Hochschule Zürich