Decision Support Systems Based on Machine Learning for Detection of Liver Diseases

The liver, the largest metabolic organ in the human body, plays a crucial role in maintaining essential functions. However, the rise in liver diseases in recent years has become a significant public health issue. There are various causes of these diseases, and the impact of technology in the health field has become necessary to analyze and manage diseases. Although there are various causes of liver diseases, analyzing and diagnosing them is crucial. Technological developments have made it possible to use machine learning methods to analyze large data sets. In this study, machine learning techniques are used in the early diagnosis and classification of liver disease, and the results are evaluated. The ‘ILPD (Indian Liver Patient Dataset)’ dataset in the UCI Machine Learning Repository has been evaluated using six modern machine learning techniques, including logistic regression, random forest, and decision trees. The study also analyses the effects of data preprocessing, feature selection, and normalization on model performance. Model results show that feature selection and normalization are essential in improving model accuracy. The Logistic Regression model achieved the best performance with an accuracy of 80%, while other algorithms, such as Random Forest and AdaBoost, also achieved high accuracy rates. This research demonstrates the potential of machine learning techniques in the early detection of liver diseases and provides a basis for future work.