Real-Time Forecast of Influenza Outbreak Using Dynamic Network Marker Based on Minimum Spanning Tree

Figure 2

The overall algorithm structure of MST-DNM method. First, model a city network based on its administrative divisions and the geographical relationship and map the corresponding clinic-visiting record matrix into the city network. Then, regard a week as a candidate tipping point, weight the city network, and calculate its minimum spanning tree’s length as the MST-DNM score . Finally, according to a logistic regression model trained by other years’ dataset, calculate the probability of belonging to 1, i.e., . If this probability is greater than or equal to 0.5, week is considered as the tipping point. Otherwise, week is classified as the normal state, and the algorithm carries on with week .