Smart Health Analytics: Evaluating ML Models in Hypothyroid Prediction

Background: The prevalence of thyroid disorders has increased in recent years. The early identification of thyroid disease is especially crucial because of its critical involvement in metabolism. Motivation: The primary function of thyroid hormones is to regulate metabolism, which converts food into energy. A healthy metabolism and the proper functioning of the heart, brain, muscles, and other organs are maintained by having enough thyroid hormones. Growth rate, energy level, heart rate, and body temperature are all regulated by a steady metabolism. Objective: To identify the appropriate machine learning algorithm in the prediction hypothyroid disorder at the earliest. With accuracy and effectiveness, artificial intelligence (AI) has the potential to improve thyroid illness early detection. A subset of machine learning (ML). Artificial intelligence (AI) has the potential to improve the accuracy and efficacy of thyroid illness early detection. As a branch of artificial intelligence, machine learning (ML) involves teaching machines to recognize patterns in data. In a number of ways, using machine learning techniques can improve the accuracy and effectiveness of thyroid disease early detection strategies. Method: By employing a variety of machine learning techniques, including Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), Logistic Regression (LR), K-Nearest Neighbour (KNN), Naïve Bayes (NB), and XGBoost, this work seeks to address these problems. Findings: The results indicate that utilizing Decision Tree and Random Forest, respectively, produced accuracy scores of 0.95 and 0.97. Novelty: In this paper various machine learning algorithms was compared and revealed that Random Forest algorithm is an efficient algorithm.