A Hierarchical Model for Association Rule Mining of Sequential Events: An Approach to Automated Medical Symptom Prediction

19 Pages Posted: 8 Jan 2011 Last revised: 30 Jan 2011

See all articles by Tyler McCormick

Tyler McCormick

Columbia University - Department of Statistics

Cynthia Rudin

Duke University; Duke University - Pratt School of Engineering

David Madigan

Columbia University - Department of Statistics

Date Written: January 6, 2011

Abstract

In many healthcare settings, patients visit healthcare professionals periodically and report multiple medical conditions, or symptoms, at each encounter. We propose a statistical modeling technique, called the Hierarchical Association Rule Model (HARM), that predicts a patient's possible future symptoms given the patient's current and past history of reported symptoms. The core of our technique is a Bayesian hierarchical model for selecting predictive association rules (such as "symptom 1 and symptom 2 → symptom 3") from a large set of candidate rules. Because this method "borrows strength" using the symptoms 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 symptoms is available.

Keywords: Hierarchical Bayesian Modeling, Association Rules, Medical Symptom Prediction

Suggested Citation

McCormick, Tyler and Rudin, Cynthia and Rudin, Cynthia and Madigan, David, A Hierarchical Model for Association Rule Mining of Sequential Events: An Approach to Automated Medical Symptom Prediction (January 6, 2011). MIT Sloan Research Paper , Available at SSRN: https://ssrn.com/abstract=1736062 or http://dx.doi.org/10.2139/ssrn.1736062

Tyler McCormick (Contact Author)

Columbia University - Department of Statistics ( email )

Mail Code 4403
New York, NY 10027
United States

Cynthia Rudin

Duke University ( email )

Department of Computer Science
LSRC Building
Durham, NC 27708-0204
United States

Duke University - Pratt School of Engineering ( email )

Durham, NC 27708
United States

David Madigan

Columbia University - Department of Statistics ( email )

Mail Code 4403
New York, NY 10027
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
367
Abstract Views
1,958
rank
102,808
PlumX Metrics