Machine Learning Approaches for Liver Disease Prediction: Performance and Model Evaluation

Liver disease continues to be a primary contributor to global mortality, requiring prompt and precise identification for effective treatments. This study assesses the efficacy of three machine learning models—Logistic Regression, Neural Network, and Naïve Bayes—in forecasting liver illness utilizing the Indian Liver Patient Dataset (ILPD). Key performance parameters, including as Area Under the Curve (AUC), Classification Accuracy (CA), F1 score, Precision, and Recall, were examined. The findings demonstrate that Neural Networks attain the highest classification accuracy (0.703), but Logistic Regression exhibits more computing efficiency, and Naïve Bayes delivers a balanced performance across several measures. This study emphasizes the balance between precision and computational economy, offering guidance for model selection tailored to specific applications in liver disease prediction.