Effective Manner of Healthcare Management System for Hepatitis virus using PCA-BFF Optimization Algorithm with Clinical Data

Hepatitis, originating from a viral bacterium, adversely affects the liver. Due to limited precautionary knowledge, the death rate is increasing. Early-stage intervention is vital for improved living standards. Machine Learning (ML) methods effectively predict diseases early, reducing human mortality. The proposed method unfolds in two phases, In the first phase, the feature detection technique employs a Principal Component Analysis (PCA) with a Hybrid Bat and FireFly Optimization Algorithm (HBFFOA) to enhance performance. In the second phase, to decrease misclassification errors and increase accuracy, the Iterative Dichotomiser with Decision Tree (ID3-DT) method is adopted. According to the experimental results, the PCA-HBFFOA technique consistently outperforms an accuracy rate of 97.1%. Evaluation metrics encompass accuracy, precision, recall, f-measure, and processing time. Additionally, the proposed method exhibits shorter execution times of 3.5 seconds for the hepatitis dataset. Looking ahead, future work will explore an ensemble classification method tailored to diverse hepatitis datasets, aiming to further optimize the overall solution.