Liver Diseases Disorder Prediction Approach Using Exploitation AI and Intelligent Ensemble Machine Learning Classifiers

Abstract

Liver disease is a worldwide health concern that greatly impacts rates of morbidity and newlinemortality. Prompt detection is crucial for efficient disease management and newlineintervention. Developing a predictive Machine Learning(ML) model to expedite prompt newlinediagnosis is of great significance. This research investigates the efficacy of several newlinemethods for machine learning(ML), such as Random Forest (RF), Light GBM (LGBM), newlineHistogram-based Gradient Boosting (HIST GRAD.), (DT) Decision Tree , Bagging newline(BAG), Artificial Neural Net. (ANN), and 1D Convolutional Neural Nets. (1D CNN), in newlinevarious predicting tasks. Additionally, it constructs three hybrid models that include newlineforecasts from these models and assesses their efficacy. The study employs a range of newlinedata pre-processing methods, such as handling outliers, imputing NULL values, newlinebalancing classes, and normalizing data, to improve performance of the model. RF is newlineutilized for the purpose of feature selection. and Principal Component Analysis used for feature engineering in order to enhance prediction accuracy. After successfully implementing the proposed methodologies, all three hybrid models have shown extraordinary accuracy. In every round of evaluation, they achieved a 97% accuracy newlinerate, as well as noteworthy gains in recall, precision, and F1 score measures.