Deliverables of Group B3 (2021) for Programming for Data Science
Urban mobility plays an important role in addressing urban livability. Traffic accidents occur every day and besides the monetary and human damage that often result from them, they also have an important impact in the urban mobility of several other people, therefore affecting the livability of the cities. Accidents often occur because of human error, however, external factors such as the condition of the road, the presence or absence of road signaling and the meteorological conditions, among others, can affect or increase their occurrence since they can produce changes in perception of the drivers. As such, the Lisbon Municipality launched a challenge that aims to identify the points with the highest incidence of road accidents and their correlation with other factors, namely signaling, orientation and inclination of roads, for 2019. In this sense, this report focuses on identifying the geographical areas with higher prevalence of accidents using geo-spatial visualization technics (GeoPandas and ArcGIS). Also, regression techniques were used to control endogenous factors of each parish, therefore allowing to understand the real risk of accident in each of them.
Please download the source data from NOVA IMS Onedrive. The file sizes were too big for github to handle.
Unzip the folder "data" to the same folder of the notebook.