<![CDATA[The liver is a vital human organ that plays a crucial role in detoxification, cholesterol regulation, and various metabolic activities within the body. Impairment of liver function can lead to several diseases such as hepatitis, liver cancer, cirrhosis, and other liver-related conditions. In Indonesia, approximately 0.6% of the population is identified as having hepatitis, despite the implementation of the HB 0–4 immunization program by the Ministry of Health. Liver disease is a common public health issue, with WHO data reporting an annual death toll of 1.2 million people due to liver-related illnesses in Southeast Asia and Africa. The importance of early detection of liver disease symptoms highlights the need for a predictive system capable of accurately identifying individuals at risk. This study employs a machine learning approach using K-Nearest Neighbor (K-NN) and Decision Tree classification algorithms, enhanced by the application of the Adaboost ensemble learning technique to optimize their performance. Evaluation results show that Adaboost improves the accuracy of the K-NN algorithm to 95.77% and the accuracy of the Decision Tree to 100%. Although the improvement in K-NN is quite significant, Adaboost does not have a substantial impact on the accuracy of the Decision Tree. This research indicates that the Adaboost method is effective in enhancing the classification performance for liver disease, particularly when applied to the K-NN algorithm.]]>
Copyrights © 2025
