A Study on the Application of Explainable AI on Ensemble Models for Predictive Analysis of Chronic Kidney Disease

This project explores the use of Explainable AI (XAI) in ensemble models for predicting Chronic Kidney Disease (CKD) using the dataset from Kaggle. The study employs various machine learning algorithms, including Logistic Regression, Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naive Bayes, and Feed-Forward Neural Networks (FNN), to build predictive models for CKD risk. Data preprocessing, feature scaling, and model training are integral to the process. Additionally, LIME (Local Interpretable Model-Agnostic Explanations) is applied to provide interpretable results, ensuring transparency in predictions. The system is deployed as a web application using HTML, CSS, JavaScript, and Flask, allowing users to input clinical data and receive predictions, along with LIME-based explanations of the results. This approach aims to improve prediction accuracy while maintaining model transparency for better decision-making.
Keywords: Chronic Kidney Disease, Explainable AI, LIME, Random Forest, Logistic Regression, SVM, KNN, Naive Bayes, Feed-Forward Neural Network, Predictive Modeling.