<![CDATA[This study aims to develop an early detection system for thyroid disease using the Naive Bayes algorithm. The dataset used is the Thyroid Disease Dataset from the UCI Machine Learning Repository, consisting of thousands of patient records. Prior to model training, the data undergoes preprocessing steps such as handling missing values, numerical normalization, and categorical encoding. The classification process involves calculating the prior probability, likelihood, and posterior probability for each class: normal, hypothyroid, and hyperthyroid. The system also presents the probability percentage for each class as an automated diagnosis result. Model accuracy is evaluated using a Confusion Matrix, achieving an accuracy score of 98.01% on the test data. These results indicate that the proposed approach can effectively and accurately classify thyroid conditions for early diagnosis purposes.]]>
Copyrights © 2025
