Predict the number of future emergency calls in different areas of Moscow using geographical data
Is it possible to predict where emergency services will be needed in the city (police/ambulance/fire department/...)?
We provide daily geodata for Moscow on the number of incoming calls to 112. In this task, it is necessary to use data from open sources, such as Open Street Map.
After training in the western half of Moscow, participants have to predict emergency calls for the eastern half. The prediction quality metric is the Kendall rank correlation coefficient (Kendall's tau). This metric focuses on the order of the predicted number of calls, but not on their exact values.
The total prize fund is 500,000 rubles! In addition to the usual top 3, judges will also award 75,000 rubles for the best public script, as well as for the best journalism project based on the provided data, which should be shared on the forum. The award will be held at Data Fest⁵. Additional special souvenirs will be granted to the top 50 participants!
The winners in the nominations for the best data journalism projects and the best public script will be determined by a jury consisting of MTS experts and competition organizers.
Your scripts and results should be publicly shared in the appropriate sections of our forum.
The results of additional nominations will be announced on April 28 at the free Data Fest conference. The official award ceremony for the winners of all categories will also be held on April 28.