ITD-ML: Improving Diagnosis Capabilities for Thyroid Disease Using Machine Learning

The human immune system is an effective, self-organizing, and distributed defense mechanism that wards off pathogens. Nevertheless, autoimmune diseases provide a significant challenge since they trigger the immune system to assault the body's own tissues and organs, resulting in inflammation and tissue damage. There is a chance that this illness might be fatal. As an example of an autoimmune illness, one of the more common forms is Hashimoto's thyroiditis, which affects the thyroid gland. There is an increase in the incidence of thyroid problems, notably hypothyroidism, among people in India, with around one in ten of them being afflicted. This condition mostly affects women, particularly those of reproductive age, and this is a problem because of the gender ratio. This research investigates the possibility of forecasting thyroid illness using data obtained from the UCI Machine Learning Repository by using machine learning. The data was obtained from the repository. To improve the accuracy of thyroid illness detection, this work makes use of a technique called Correlation-Based Feature Selection (CFS), as well as other multi-class classification approaches, such as Decision Tree, k-nearest neighbor, and Support Vector Classifier.