A Model for Predicting the Class of Illicit Drug Suspects and Offenders

• Published date: 01 April 2022
• Author: Oluwafemi S. Balogun,Temidayo O. Omotehinwa,Richard O. Agjei,Toluwalase J. Akingbade,Donald D. Atsa'am,Samuel N. O. Devine
• DOI: 10.1177/00220426211049358
• Date: 01 April 2022
• Subject Matter: Articles

Article

Journal of Drug Issues

2022, Vol. 52(2) 168–181

© The Author(s) 2021

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/00220426211049358

journals.sagepub.com/home/jod

A Model for Predicting the

Class of Illicit Drug Suspects

and Offenders

Donald D. Atsa’am

1

, Oluwafemi S. Balogun

2

, Richard O. Agjei

3

,

Samuel N. O. Devine

4

, Toluwalase J. Akingbade

5

, and

Temidayo O. Omotehinwa

6



Abstract

In this study, the artif‌icial neural network was deployed to develop a classif‌ication model for

predicting the class of a drug-related suspect into either the drug peddler or non-drug peddler

class. A dataset consisting of 262 observations on drug suspects and offenders in central Nigeria

was used to train the model which uses parameters such as exhibit type, suspect’s age, exhibit

weight, and suspect’s gender to predict the class of a suspect, with a predictive accuracy of 83%.

The model sets the pace for the implementation of a full system for use at airports, seaports,

police stations, and by security agents concerned with drug-related matters. The accurate

classif‌ication of suspects and offenders will ensure a faster and correct reference to the sections of

the drug law that correspond to a particular offence for appropriate actions such as prosecution

or rehabilitation.

Keywords

drug use, drug traff‌icking, suspect classif‌ication, classif‌ication model, artif‌icial neural network

Introduction

Drug abuse and drug traff‌icking remain public health challenges globally. According to the United

Nations Off‌ice on Drugs and Crime (UNODC), approximately 271 million (15–64 years) people

globally had used drugs in 2018 as evidenced in the 2019 report. It is estimated that 35 million

1

Department of Computer Science and Informatics, Faculty of Natural and Agricultural Sciences, University of the Free

State, Bloemfontein, South Africa

2

School of Computing, University of Eastern Finland, Kuopio, Finland

3

Department of Public Health, University of Central Nicaragua Medical Center, Semaforos del Zumen, Nicaragua

4

Department of Information and Communication Technology, Presbyterian University College, Abetif‌i-Kwahu, Ghana

5

Department of Mathematical Sciences, Kogi State University, Anyigba, Nigeria

6

Department of Mathematical Sciences, Achievers University, Owo, Nigeria

Corresponding Author:

Donald D. Atsa’am, Department of Computer Science and Informatics, Faculty of Natural and Agricultural Sciences,

University of the Free State, Qwaqwa Campus, Phuthaditjhaba 9866, South Africa.

Email: donatsaam@alumni.emu.edu.tr