FO_RF_LG - A Hybrid Fruit-Fly Optimization - Random Forest with LightGBM based Classification Model for Infectious Disease Prediction in Big Data Environment

Infectious disease prediction plays a crucial role in early diagnosis, outbreak control, and patient management. However, traditional machine learning models face challenges in handling large-scale healthcare data, feature redundancy, and computational efficiency. This paper proposes a hybrid machine learning model that integrates Fruit Fly Optimization (FO) and Random Forest (RF) for feature selection, followed by LightGBM (LG) for classification (FO_RF_LG). The FO_RF method efficiently selects the most relevant disease-related features, reducing dimensionality while maintaining predictive accuracy. The optimized feature set is then classified using LightGBM (LG), a gradient boosting framework known for its speed and accuracy. The proposed model was evaluated on multiple infectious disease datasets, including tuberculosis, and influenza, demonstrating higher accuracy. Compared to traditional classifiers such as Logistic Regression, SVM, and standard RF, the FO_RF_LG model achieved superior performance with huge reduction in training time and higher improvement in big data processing efficiency. These results highlight the effectiveness of the proposed hybrid model for real-time infectious disease prediction in large-scale healthcare environments.