Datasets

Where is it?

The task is to predict the movement of 3000 individuals each in cities B, C, and D, during days 61 to 75, using movement data of individuals in city A (100,000 individuals’ full trajectories from day 1 to 75) and cities B, C, D, which includes a total of 25,000, 20,000, and 6,000 individuals, respectively. The locations (x,y) that need to be predicted are masked with cell numbers ‘999’. 

FYI: the human mobility dataset is a similar format as the dataset used in last year’s HuMob Challenge. The dataset can be found here: https://zenodo.org/records/10836269 

The data descriptor was published in Nature Scientific Data. Check it out here: https://www.nature.com/articles/s41597-024-03237-9

Multi-City Human Mobility Prediction