<![CDATA[Chronic Kidney Disease (CKD) is a critical global health issue, characterized by significant morbidity and mortality. Early detection is vital for effective management and improved patient outcomes. This study explores the application of the Gaussian Naive Bayes algorithm to predict CKD using a comprehensive dataset from Kaggle, comprising health information from 1,659 patients. The research involves detailed data pre-processing, including feature selection, data scaling, and an 80/20 split for training and testing. The model's performance was evaluated using 5-fold cross-validation, resulting in an average accuracy of 89.93%, precision of 88.15%, recall of 89.93%, and F1-score of 88.42%. These metrics highlight the model's robustness and reliability in identifying CKD cases. Visualizations such as correlation heatmaps, 3D PCA, and t-SNE plots were used to understand feature relationships and data distribution. The results confirm the hypothesis that Gaussian Naive Bayes can effectively predict CKD, providing a reliable tool for early diagnosis. This study contributes to the medical field by demonstrating the utility of machine learning in improving diagnostic accuracy. However, limitations such as dataset biases and the need for comparison with other algorithms are acknowledged. Future research should focus on expanding the dataset, incorporating more features, and exploring additional machine learning models to enhance predictive performance and generalizability. Practical implications suggest that integrating such models into clinical practice could significantly improve patient management and outcomes.]]>
Copyrights © 2024
