An Improved Supervised Classification Algorithm in Healthcare Diagnostics for Predicting Opioid Habit Disorder

Abstract

Opioid Habit Disorder (OHD), which has become a mass health epidemic, is defined as the psychological or physical dependency on opioids. This study demonstrates how supervised machine learning procedures help us investigate and examine massive data to discover the hidden patterns in any disease to deliver adapted dealing and predict the disease in any patient. This work presents a generalized model for forecasting a disease in the healthcare sector. The proposed model was investigated and tested using a reduced feature-set of the Opioid Habit Disorder (OHD) dataset collected from the National Survey on Drug Use and Health (NSDUH) using an improved Iterative Dichotomiser 3 (pro-IDT) algorithm. The proposed healthcare model is also compared with further machine learning algorithms such as ID3, Random Forest, and Bayesian Classifier in Python programming. The performance of the proposed work and other machine-learning algorithms has estimated accuracy, precision, misclassification rate, recall, specificity, and F1 score.