Data mining is one of the most promising areas of research that has become increasingly popular in health care. The objective of this research article is to elucidate a transdisciplinary approach to classify thyroid levels in patients using data mining techniques. The data set consisting of more than 2 thyroid conditions along with the normal values listed 21 values. The classifier chosen carefully to get the optimized accuracy and falls in different classification category namely J48, random forest, and random tree from tree, decision table from rules, multilayer perceptron from functions, naïve Bayesian from Bayes, and AdaBoost from meta respectively. The J48 classifier displays tree that will assist with better interpretation based on the values and helps to easily determine new value combinations. The Thyropred System graphic is finally presented, which guides the diagnosis of thyroid disease. Thorough consultation with experts along with this prediction system guides the decision of further medication. The J48 classifier provides the best accuracy when compared with the other tested classifiers. © 2019 ASTM International. All rights reserved.