A Hybrid Approach for Liver Disease Diagnosis and Prognosis

The process of finding patterns in massive datasets to inform decisions is known as machine learning. Machines can learn through supervised, unsupervised, semi-supervised, or reinforced processes. The UCI Repository's Liver Patient data set (Supervised Learning) is used in this work. Data from hospital medical examinations on liver patients can help improve their conditions in the future. Algorithms and classifiers use historical and classified patient data to predict future outcomes. The algorithms Bagging, Random Forest, Decision Tree, XGBoost, AdaBoost, and LightGBM are used to forecast liver patients. The algorithm achieved high accuracy after feature selection, as determined by analysis and calculations