Enhancing Liver Disease Diagnosis using Ensemble Machine Learning: A Comparative Analysis of Boosting, Bagging and Voting Classifiers

Liver disease is a serious issue that is spread across the world wide, it requires a very instant and meticulous diagnosis in order to cure early and effectively. Conventional diagnostic methods mostly are dependent on common machine learning models and clinical competence, they both may not have good predictive power. The proposed work shows how accurately ensemble learning methods – Bagging-RF, Boosting-GB and Voting Classifiers improve the diagnosis of liver unwellness. Based on essential measures like accuracy, precision, recall, and F1-score, compare the performance of various ensemble algorithms using real-world medical datasets. The results show the use of a multiple ensemble technique to enhance diagnostic reliability by outperforming individual classifiers. While Bagging and Voting improve the strength of the model, reduce the overfitting, Boosting has the best ability to predict among the assessed techniques and perform well across all the assessment measures. This study shows how effectively group learning in medical prognosis can be providing a successful and accurate strategy.