Accident Hotspot Prediction

MoveHack - Global Mobility Hackathon conducted by NITI AAYOG, Government of India

Problem Statement

Road safety is of paramount importance for the Government and the entire mobility ecosystem. While the number of road accidents has reduced over the last couple of years, still the overall number of accidents that occurred in 2017 is alarming. It has been assessed that 90% of the accident fatalities occur due to human error. “Brasilia Declaration on Road Safety” envisages to halve road traffic deaths by the end of this decade. There are number of interventions being adopted to reduce the occurrence of the road accidents. A technological solution/platform which can rate driver behavior and predict accident hot spot is the need of the hour. This will enable enforcement agencies and the commuters to take preemptive actions to avoid accidents.

Team Members

1. Ankur Sinha, India
2. Aysun Onalan, Turkey
3. Berkay Kopru, Turkey
4. Jie Fu, China

Instructions

Run the prediction_model.ipynb file to perform the analysis and prediction for accident zones of Bengaluru