A Novel Approach for Prediction of Liver Disease Using Voting Algorithm Based on Machine Learning Models

Chronic liver disease, which affects a lot of people, is the leading cause of death around the world. This situation happens because of several things. Conventional methods can be used to find out what is wrong with the liver, but they can be expensive. Machine learning, which is an important part of the technological change in health care, is a key part of being able to predict diseases early on. Using feature selection and classification methods, it is possible to make software that can correctly predict liver disease. In other words, different systems get information about past patients and information about what the patients do. Classifiers can be used to guess what will happen in the future. Logistic regression, random forests, support vector classifiers, decision tree classifier, gradient boosting, K nearest neighbor classifiers, XGBoost classifiers, extra tree classifier, voting ensemble methods, and K nearest neighbor classifiers were all used to identify liver cases. Most accurate was the voting ensemble model with a rate of 97.62%.