iReMS: an intelligent recommendation system for symptom-based malaria diagnosis

This thesis addresses an important issue in early-stage malaria diagnosis by developing an intelligent recommendation system for symptom-based malaria diagnosis. Equipped with malaria classifiers built based on actual medical records using machine learning techniques, the system makes customised recommendations based on users’ profile and symptoms. With progressive user interface design, the system requires users to only input symptoms needed to make recommendations. These developments represent original contributions to malaria diagnosis and intelligent decision support research with both methodological and practical significance. The outcome of this research addresses knowledge gaps in malaria symptoms and contributes to Indonesia’s health sector.