Ensemble Learning for Improved Thyroid Disease Prediction: A Voting Classifier Approach

Thyroid disease represents a significant contributor to challenges in both medical diagnosis and the prediction of its onset, making it a complex area of study within medical research. This research thoroughly analyses the effective use of machine-learning approaches for thyroid identification. In our research, we use a Voting Classifier using five machine learning models: Support Vector Machine, Logistic Regression, Random Forest, Decision tree, and Gradient Boosting Classifier to detect negative patients or hyperthyroid and hypothyroid patients. The combined result of multiple machine learning models refers to 99.6% accuracy which surpasses the previous results that are mostly single machine learning models. The findings indicate that when considering both accuracy and computational complexity, machine learning models emerge as a preferable option for detecting thyroid disease.