Each submission should include a demonstration of the metric on at least one data set. We provide two example data-sets here, but competitors are welcome to use or create any test data they like, as long as it includes both time-segment and map-segment information and is publicly available.
The first data set we've provided contains event data, the 2019 Baltimore, MD 911-Call and Police Incident data, which is being used as the development phase data set for the first Sprint of the Differential Privacy Temporal Map Challenge (https://deid.drivendata.org/). The second data set is survey data, including demographic and financial features-- A small subset of IPUMS American Community Survey data for Maryland, from 2010-2018. Both examples include ground truth data, as well as privatized data at varying levels of quality, and data dictionaries.
Our leadership team is thrilled with the results! We were able to successfully identify three winning solutions. The diversity of the innovators was so refreshing, we get how our solutions can literally live anywhere in the world.