A Comparitive Study on Liver Disease Prediction using Supervised Learning Algorithms with Hyperparameter Tuning

Abstract

The World Health Organization estimates that liver disease or liver cancer results in one million deaths annually, and that ten new cases of hepatitis B and C are diagnosed each day. Given the challenges and expenses associated with diagnosing liver disease, this study aims to evaluate the effectiveness of a range of Supervised Machine Learning algorithms in predicting and detecting the disease, with the objective of reducing healthcare costs. To achieve this goal, the Indian Liver Patient Dataset from UCI (University of California Irvine) repository is used. The study employs various classification algorithms, including Random Forest, Decision Tree, Decision Tree SMOTE, Support Vector Classifier, K-Nearest Neighbour, AdaBoost, Stochastic Gradient Descent, and Artificial Neural Network. The study shows that ANN is the most effective algorithm, achieving an impressive margin of 87 percent over the other algorithms.