Abstract

In recent decades, thyroid dysfunction has become a widespread illness that affects millions of individuals worldwide, mostly women between the ages of 17 to 54. TSH (Thyroid-Stimulating Hormone) that are too high or too low may be a sign of a thyroid problem. The extreme stage of thyroid results in heart problem, depression, etc. Here implements the proactive system to predict the thyroid at its earliest stage is done. This will reduce the death rate and other side effects due to thyroid problems. The techniques used in this work include logistic regression, KNN (k-nearest neighbors) and Decision trees, and these was selected for its different method. These algorithms are the best and most suitable to deal with the prediction of thyroid disease at the earliest stage with less complexity and more accuracy in the implementation. Based on the results obtained, the logistic regression is better and, hence used for the problem in the thyroid disease.