What it does
Predicts with a 70% accuracy what the category of the call (the topic of conversation) is, given the date-time, weather, geographical location of the source call and the City of Montreal department receiving the call.
How I built it
Challenges I ran into
Unclean data Very unclean data A lot of data Less Compute Power. Google Colab wasn't a very great experience tbh.
Accomplishments that I'm proud of
Build a model that predicts decent!
What I learned
A lot about the AI space in Montreal, some of the exciting works some of the startups are doing and, in general, brushed up my ML which I hadn't touched in a while.
What's next for Predicting the reason for their call
1) Using calls data with crime data (use a reverse_geocoder to get the Arrondisments) 2) Using calls data with income data A lot of interesting things can be done with these 3.