Date of Graduation

12-2019

Document Type

Thesis

Degree Name

Master of Science in Industrial Engineering (MSIE)

Degree Level

Graduate

Department

Industrial Engineering

Advisor/Mentor

Chaovalitwongse, W. Art

Committee Member

Zhang, Shengfan

Second Committee Member

Martin, Bradley C.

Third Committee Member

Rainwater, Chase E.

Keywords

Deep learning; Feature engineering; Machine learning; Medical claims; Opioid overdose; Prediction

Abstract

The goal of this project is to develop an efficient methodology for extracting features from time-dependent variables in transaction data. Transaction data is collected at varying time intervals making feature extraction more difficult. Unsupervised representational learning techniques are investigated, and the results compared with those from other feature engineering techniques. A successful methodology provides features that improve the accuracy of any machine learning technique. This methodology is then applied to insurance claims data in order to find features to predict whether a patient is at risk of overdosing on opioids. This data covers prescription, inpatient, and outpatient transactions. Features created are input to recurrent neural networks with long short-term memory cells. Hyperparameters are found through Bayesian optimization. Validation data features are reduced using weights from the best model and compared against those found using unsupervised learning techniques in other classifiers.

Download

Share

COinS