Performance Evaluation of Machine Learning Methods for Thyroid Prediction

Diagnosis of the disease is a crucial stage as it leads to the patient’s proper treatment. Thyroid disease is one of the most chronic endocrine illnesses. The thyroid gland becomes dysfunctional as a result of thyroid disease. Thyroid gland secretes the hormones which controls the metabolism of the body. There are mainly two thyroid disorders namely hypothyroid and hyperthyroid. Using machine learning techniques including Random Forest, Naive Bayes, and Decision Tree, this study aims to detect thyroid disorders. Thyroid dataset is collected from Lata Mangeshkar Hospital at Nagpur. The dataset consist of total 225 patient’s data. The performance of the algorithms was evaluated using different metrics as accuracy, precision, recall and f-score.