Liver cirrhosis is a chronic liver disease which causes the severe complications that include liver failure and much more. Accurate diagnosis of disease is needed for managing it that further improves the patient liver performance. There are many traditional methods available for estimating the progression and survival rate of cirrhosis which commonly include clinical scoring systems such as Child-Pugh and MELD scores which are informative but doesn’t capture the full analysis of the each cases. The aim of this work is to define cirrhosis through clinical data, on various outcomes. Using a database of patients who are diagnosed with liver disease, we applied statistical machine learning techniques to develop a predictive model that can assess the patient outcomes like Age, sex, and what are the causes of liver disease. The model was trained on 80% of the data set, validated on 20%, and gave a result of 90% with a validation accuracy, illustrating that it effectively predicts survivability and separates stages from each other regarding liver conditions. The F1 measure was able to point to the balance of precision to recall of the model, pointing to its proper identification for severe cases without misclassification. These measures establish and support the model’s effectiveness in clinical decision-making-the model supports early interventions of improved patient outcomes. Using a lively dataset of cirrhosis patients, the model is developed and evaluated several ML models, including decision trees, random forests, and support vector machines. Each model was evaluated based on the performance metrics like accuracy, sensitivity etc. Our study shows that early recognition of the disease will consequently impact the outcomes of the survivals.