Enhanced Hypothyroidism Prediction Through Meta-Learning Approaches using Machine Learning.

These days, thyroid disease is the most prevalent illness that can cause a number of different health problems. Thyroid hormone disorders, which can result in issues like hypothyroidism or hyperthyroidism, are the primary cause of this illness. It is essential yet difficult to foresee the issue well because it is frequently underestimated. Thyroid tests based on many factors to identify thyroid hormone behaviour are the conventional method of diagnosing hypothyroidism. Algorithms for machine learning are crucial for early disease detection and prediction. To determine the most efficient and precise predictions, machine learning classifiers can be used to create a variety of studies and models. Our research attempts to apply various machine-learning classification methods and create a model for hypothyroidism prediction. In order to determine which classifier is optimal for constructing a classification model, we also compare the machine learning methods. In order to achieve the best outcomes and increase accuracy, we also want to employ ensemble learning approaches. The best-performing models when combined with the boosting approach are the Adaboost classifier and gradient boosting, which yield the greatest accuracy of 97.57% and 97.51%, respectively. We have reached 97.23% with gradient boosting. Additionally, we have put in place a web program that forecasts hypothyroidism.