deletions | additions       diff --git a/Introduction.tex b/Introduction.tex  index 5911024..4eae862 100644  --- a/Introduction.tex  +++ b/Introduction.tex 
...
  This data-driven methodology is applied to a large video dataset collected at more than 20 roundabouts in Quebec to study road user behaviour and their safety. More than 40 camera views define roundabout sub-regions delimited by an entry and the following exit, constituting a weaving zone with the vehicles already within the roundabout. Each camera recorded 12 to 16 h of video on a given day, which constitutes a dataset with 600 h of video data. Applying the proposed method to this large dataset yields considerable amounts of indicators, from individual road user measurements, e.g. speed, to individual interaction measurements, e.g. TTC, to aggregated indicators per road user or interaction, to aggregated indicators per site over time and space.  Analyzing such big data is a challenge of a magnitude that has never been undertaken before in road safety research. It holds the key to understanding the processes that lead road user to collide, to design and validate safety indicators that do not require to wait for accidents to occur. The approach will be demonstrated on this video dataset to identify roundabout characteristics that influence road \subsubsection{Size of data (hours, GB, framerate, resolution)}