Integrating Semantic Literature Review and Simple Experiments: Comparison of AI Algorithms in Tropical Disease Case Studies

Abstract

Tropical diseases represent a global health challenge requiring innovative approaches for diagnosis and prediction. This study integrates semantic literature review with comparative experiments of artificial intelligence algorithms to analyze the effectiveness of various AI methods in handling tropical disease case studies. Through systematic analysis of 127 scientific publications from 2018-2025, we identified trends in using machine learning, deep learning, and natural language processing algorithms in tropical health domains. Comparative experiments were conducted on five main algorithms: Support Vector Machine (SVM), Random Forest, Neural Network, Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) using datasets of malaria, dengue, and tuberculosis. Results show that CNN achieved the highest accuracy of 94.7% for malaria microscopic image classification, while LSTM performed best for dengue outbreak prediction with an F1-score of 0.89 using temporal data. Random Forest showed the most consistent performance across different diseases with an average accuracy of 87.3%. Semantic analysis revealed research gaps in multi-modal data integration and personalized medicine for tropical diseases. This study provides a comprehensive roadmap for AI implementation in tropical countries' healthcare systems.