Bayesian hierarchical rule modeling for predicting medical conditions

Published

Journal Article

We propose a statistical modeling technique, called the Hierarchical Association Rule Model (HARM), that predicts a patient's possible future medical conditions given the patient's current and past history of reported conditions. The core of our technique is a Bayesian hierarchical model for selecting predictive association rules (such as "condition 1 and condition 2 → condition 3") from a large set of candidate rules. Because this method "borrows strength" using the conditions of many similar patients, it is able to provide predictions specialized to any given patient, even when little information about the patient's history of conditions is available. © Institute of Mathematical Statistics, 2012.

Full Text

Duke Authors

Cited Authors

  • McCormick, TH; Rudin, C; Madigan, D

Published Date

  • June 1, 2012

Published In

Volume / Issue

  • 6 / 2

Start / End Page

  • 652 - 668

Electronic International Standard Serial Number (EISSN)

  • 1941-7330

International Standard Serial Number (ISSN)

  • 1932-6157

Digital Object Identifier (DOI)

  • 10.1214/11-AOAS522

Citation Source

  • Scopus