Accurate prediction of liver disease is crucial for effective patient management and treatment. This research integrates various machine learning (ML) models with Explainable Artificial Intelligence (XAI) techniques to enhance the diagnosis of liver disease using two comprehensive healthcare datasets: the Indian Liver Patient dataset and the Cirrhosis dataset. We employ ML models such as logistic regression, K-nearest neighbors, decision trees, support vector machines, random forests, and XGBoost to predict liver disease. Additionally, we apply XAI techniques including SHAP, LIME, and Breakdown to interpret the predictions of these models. Our results demonstrate the efficacy of these ML models in accurately diagnosing liver disease and emphasize the critical role of transparency and interpretability in clinical applications. By elucidating the key features driving model predictions, our approach aims to support healthcare professionals in making informed decisions, ultimately improving patient outcomes.