Neural Network-Based Models For Predicting Healthcare Needs In International Travel Coverage Plans

Abstract

Safeguarding health from diseases during international travel is essential. International travel health insurance is a type of risk management that offers coverage for health issues, such as disease occurrences, to which travelers could be exposed at their destination. Offering proper international travel coverage plans during trip planning assists travelers in mitigating such health issues since a traveler can focus on the planned trip instead of worrying. This research developed and tested deep learning algorithms trained on large, vast dimensional, long historical periods, and near-real-time multi-feature healthcare data about the final destination of a traveler.

In the past, medical experts utilized various algorithmic models based on traditional statistical analysis and simple classification and regression techniques for predicting travel health risk issues without giving much attention to deviations from randomness, accuracy, and overfitting of the selected models, which have subsequently been labeled as health-tuned risk prediction systems. Recently, data-driven classification and prediction algorithmic models are starting to more adaptively classify, compare, and predict travel health risk issues. This research contributes to the existing family of data-driven medical tourism research with in-depth data analytical experiments using state-of-the-art industry-discovered techniques adopted for classification and forecasting tasks.