Abstract
Risk assessment of public health emergencies plays a crucial role in timely response and effective management of pandemics. Traditional risk assessment methods often rely on a single data source, neglecting the integration of multidimensional information, which limits the accuracy and comprehensiveness of the evaluation. This paper proposes a risk assessment method based on multi-source data fusion, integrating the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method and a hierarchical filtering technique, applied to the risk assessment of public health emergencies. Initially, data from multiple sources encompassing epidemic transmission rates, population density, healthcare resources, and environmental factors are gathered from diverse fields to establish a comprehensive informational foundation for conducting risk assessments. Hierarchical filtering integrates multi-source data by reducing redundancy and improving quality. Next, the TOPSIS method is employed for comprehensive evaluation, calculating the weights and relative distances of various evaluation factors to determine the risk level of the event. Experimental results show that this method significantly enhances the accuracy of risk assessment for public health emergencies and provides a scientific basis for emergency decision-making. This research offers new insights and technical approaches for the development of risk assessment and decision support systems in the public health sector.