A Predictive Analytics-Based Decision Support System for Drug Courts

Hamed M. Zolbanin, Dursun Delen, Durand Crosby, David Wright

Research output: Contribution to journalArticle

Abstract

This study employs predictive analytics to develop a decision support system for the prediction of recidivism in drug courts. Based on the input from subject matter experts, recidivism is defined as the violation of the treatment program requirements within three years after admission. We use two data processing methods to improve the accuracy of predictions: synthetic minority oversampling and survival data mining. The former creates a balanced data set and the latter boosts the model’s performance by adding several new, informative variables to the data set. After running several tree-based machine learning algorithms on the input data, random forest achieved the best performance (AUROC = 0.884, accuracy = 80.76%). Compared with the original data, oversampling and survival data mining increased AUROC by 0.068 and 0.018, respectively. Their combined contribution to AUROC was 0.088. We present a simplified version of decision rules and explain how the decision support system can be deployed. Therefore, this paper contributes to the analytics literature by illustrating how date/time variables - in applications where the response variable is defined as the occurrence of some event within a certain period - can be used in data management to improve the performance of predictive models and the resulting decision support systems.

Original languageEnglish
JournalInformation Systems Frontiers
DOIs
StateAccepted/In press - 1 Jan 2019
Externally publishedYes

Fingerprint

Decision Support Systems
Decision support systems
Drugs
Oversampling
Data mining
Survival Data
Data Mining
Data Processing Methods
Information management
Learning algorithms
Learning systems
Random Forest
Prediction
Predictive Model
Performance Model
Decision Rules
Data Management
Date
Learning Algorithm
Machine Learning

Keywords

  • Drug court
  • Machine learning
  • Predictive analytics
  • Survival data mining

Cite this

@article{36600f7e0dc04385b9142a722cb2ed08,
title = "A Predictive Analytics-Based Decision Support System for Drug Courts",
abstract = "This study employs predictive analytics to develop a decision support system for the prediction of recidivism in drug courts. Based on the input from subject matter experts, recidivism is defined as the violation of the treatment program requirements within three years after admission. We use two data processing methods to improve the accuracy of predictions: synthetic minority oversampling and survival data mining. The former creates a balanced data set and the latter boosts the model’s performance by adding several new, informative variables to the data set. After running several tree-based machine learning algorithms on the input data, random forest achieved the best performance (AUROC = 0.884, accuracy = 80.76{\%}). Compared with the original data, oversampling and survival data mining increased AUROC by 0.068 and 0.018, respectively. Their combined contribution to AUROC was 0.088. We present a simplified version of decision rules and explain how the decision support system can be deployed. Therefore, this paper contributes to the analytics literature by illustrating how date/time variables - in applications where the response variable is defined as the occurrence of some event within a certain period - can be used in data management to improve the performance of predictive models and the resulting decision support systems.",
keywords = "Drug court, Machine learning, Predictive analytics, Survival data mining",
author = "Zolbanin, {Hamed M.} and Dursun Delen and Durand Crosby and David Wright",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s10796-019-09934-w",
language = "English",
journal = "Information Systems Frontiers",
issn = "1387-3326",
publisher = "Springer Netherlands",

}

A Predictive Analytics-Based Decision Support System for Drug Courts. / Zolbanin, Hamed M.; Delen, Dursun; Crosby, Durand; Wright, David.

In: Information Systems Frontiers, 01.01.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Predictive Analytics-Based Decision Support System for Drug Courts

AU - Zolbanin, Hamed M.

AU - Delen, Dursun

AU - Crosby, Durand

AU - Wright, David

PY - 2019/1/1

Y1 - 2019/1/1

N2 - This study employs predictive analytics to develop a decision support system for the prediction of recidivism in drug courts. Based on the input from subject matter experts, recidivism is defined as the violation of the treatment program requirements within three years after admission. We use two data processing methods to improve the accuracy of predictions: synthetic minority oversampling and survival data mining. The former creates a balanced data set and the latter boosts the model’s performance by adding several new, informative variables to the data set. After running several tree-based machine learning algorithms on the input data, random forest achieved the best performance (AUROC = 0.884, accuracy = 80.76%). Compared with the original data, oversampling and survival data mining increased AUROC by 0.068 and 0.018, respectively. Their combined contribution to AUROC was 0.088. We present a simplified version of decision rules and explain how the decision support system can be deployed. Therefore, this paper contributes to the analytics literature by illustrating how date/time variables - in applications where the response variable is defined as the occurrence of some event within a certain period - can be used in data management to improve the performance of predictive models and the resulting decision support systems.

AB - This study employs predictive analytics to develop a decision support system for the prediction of recidivism in drug courts. Based on the input from subject matter experts, recidivism is defined as the violation of the treatment program requirements within three years after admission. We use two data processing methods to improve the accuracy of predictions: synthetic minority oversampling and survival data mining. The former creates a balanced data set and the latter boosts the model’s performance by adding several new, informative variables to the data set. After running several tree-based machine learning algorithms on the input data, random forest achieved the best performance (AUROC = 0.884, accuracy = 80.76%). Compared with the original data, oversampling and survival data mining increased AUROC by 0.068 and 0.018, respectively. Their combined contribution to AUROC was 0.088. We present a simplified version of decision rules and explain how the decision support system can be deployed. Therefore, this paper contributes to the analytics literature by illustrating how date/time variables - in applications where the response variable is defined as the occurrence of some event within a certain period - can be used in data management to improve the performance of predictive models and the resulting decision support systems.

KW - Drug court

KW - Machine learning

KW - Predictive analytics

KW - Survival data mining

UR - http://www.scopus.com/inward/record.url?scp=85068030040&partnerID=8YFLogxK

U2 - 10.1007/s10796-019-09934-w

DO - 10.1007/s10796-019-09934-w

M3 - Article

AN - SCOPUS:85068030040

JO - Information Systems Frontiers

JF - Information Systems Frontiers

SN - 1387-3326

ER -