Abstract
Over seventy percent of Americans take at least one form of prescription medication, with twenty percent taking more than five. The numbers emphasize how important it is for clinicians to understand the effects of the medication and whether these medications are effective. In this paper we propose a data driven framework to predict the effectiveness of medication on a patient, specifically in the case of diabetes. Our dataset contains claims data from 1.5 million patients. A heuristic was established to evaluate the 'effectiveness' of Metformin using a set of three criteria. Decision trees and random forests were used to create prediction models on the training data and select features. The model was able to correctly predict whether a patient responded well to the medication with approximately 80% accuracy and an F1-measure of approximately 90%.
Original language | English (US) |
---|---|
Title of host publication | IEEE International Conference on Data Mining Workshops, ICDMW |
Publisher | IEEE Computer Society |
Pages | 1185-1188 |
Number of pages | 4 |
Volume | 2015-January |
Edition | January |
DOIs | |
State | Published - Jan 26 2015 |
Event | 14th IEEE International Conference on Data Mining Workshops, ICDMW 2014 - Shenzhen, China Duration: Dec 14 2014 → … |
Other
Other | 14th IEEE International Conference on Data Mining Workshops, ICDMW 2014 |
---|---|
Country/Territory | China |
City | Shenzhen |
Period | 12/14/14 → … |
Keywords
- Decision support systems
- electronic medical records
- fuzzy logic
- Medical information systems
- predictive models
- supervised learning
- support vector machines
ASJC Scopus subject areas
- Computer Science Applications
- Software