Bayesian Rule Sets for Interpretable Classification

Tong Wang; Cynthia Rudin; Finale Velez-Doshi; Yimin Liu; Erica Klampfl; Perry MacNeille

doi:10.1109/ICDM.2016.0171

Back

Conference proceeding

Bayesian Rule Sets for Interpretable Classification

Tong Wang, Cynthia Rudin, Finale Velez-Doshi, Yimin Liu, Erica Klampfl and Perry MacNeille

2016 IEEE 16th International Conference on Data Mining (ICDM), pp.1269-1274

12/2016

DOI: 10.1109/ICDM.2016.0171

View Online

Abstract

A Rule Set model consists of a small number of short rules for interpretable classification, where an instance is classified as positive if it satisfies at least one of the rules. The rule set provides reasons for predictions, and also descriptions of a particular class. We present a Bayesian framework for learning Rule Set models, with prior parameters that the user can set to encourage the model to have a desired size and shape in order to conform with a domain-specific definition of interpretability. We use an efficient inference approach for searching for the MAP solution and provide theoretical bounds to reduce computation. We apply Rule Set models to ten UCI data sets and compare the performance with other interpretable and non-interpretable models.

Computational modeling

Bayesian modeling

Simulated annealing

Predictive models

association rules

classifier

Search problems

Data models

Bayes methods

Data mining

interpretable machine learning

Details

Title: Subtitle: Bayesian Rule Sets for Interpretable Classification
Creators: Tong Wang
Cynthia Rudin
Finale Velez-Doshi
Yimin Liu
Erica Klampfl
Perry MacNeille
Resource Type: Conference proceeding
Publication Details: 2016 IEEE 16th International Conference on Data Mining (ICDM), pp.1269-1274
DOI: 10.1109/ICDM.2016.0171
eISSN: 2374-8486
Publisher: IEEE
Language: English
Date published: 12/2016
Academic Unit: Business Analytics
Record Identifier: 9984083247302771

Metrics

33 Record Views

See more details