Abstract:
Thyroid disease is rising fast among individuals these days and it affects more in women than males. Thyroid disorders are most commonly caused by abnormal thyroid hormone production. Hyperthyroidism is a condition caused by an excess of hormones. Insufficient hormone production causes hypothyroidism. Tiredness, dry skin, cold intolerance, facial swelling, menstrual cycles, and hair loss are all symptoms of thyroid dysfunction. It is far more vital to prevent rather than cure such disorders, because most therapies involve long-term medication. As a result, it's critical to look at the thyroid dataset for early disease detection so that steps can be done to avoid the deadly condition of thyroid disease. Machine learning techniques are important in the medical profession because they help doctors make better decisions, diagnose diseases more accurately, and save patients money and time. To make the data rudimentary enough for analytics to highlight the likelihood of patients developing thyroid disease, data cleaning techniques were used. The purpose of this study is to compare the accuracy and other performance criteria of multiple machine learning algorithms for predicting thyroid disease. This paper focuses on the current machine learning techniques utilized in the diagnosis and prediction of thyroid detection.
