An Effective Machine Learning Framework for Hepatitis C Virus Prediction

The deadly hepatitis C virus (HCV) is extremely prevalent and especially not treated quickly, may cause to permanent liver failure, cirrhosis or death. The establishment of prediction models through the application of machine learning techniques is hence effective. The purpose of this investigation is to find efficient machine learning framework that may utilize health data to identify people having HCV. We performed preprocessing consisting of removing null values, standardizing data using normalization and label encoding process. Although the HCV dataset was imbalanced the synthetic minority oversampling technique (SMOTE) was utilized to produce a better balanced dataset. By utilizing feature selection techniques, we also identified and ranked an assortment of risk factors. The several techniques encompassed the decision tree (DT), random forest (RF), K-nearest neighbors (KNN) and Multilayer perceptron (MLP) algorithm. The effectiveness of the models is assessed across various splitting ratios of 50:50 60:40 70:30 and 80:20 respectively. This models performance demonstrate that RF, KNN obtained 99.56%, 98.44% and DT, MLP achieved 98.67% accuracy respectively. Our experiments demonstrates exceptional accuracy of 99.56% for RF respectively. The findings may help healthcare professionals in enhancing clinical outcomes by applying a dependable HCV prediction models.