This repo deals with the projects we made during team Serendipity's venture to the National Finals of NASA SpaceApps Challenge, India
For the State Level finals, we made a web-app on Django server. The web app had used a deep learning algorithm along with KNN mapping, to inherit certain features from the dataset.
The dataset consisted of accident data from various cities in the United States, we narrowed down our searches to only those accidents, occurring in the geographical coordinates of the city that we were concerned about, namely Los Angeles.
The dataset consisted of the data regarding the particulars of the road accidents, the date, time, place and cause of the accident.
We used a deep learning model to identify the hotspots for the accidents, and used a notification based warning system that took in the email of the users who registered for our service and provided and warned them whenever they came near a accidental hotspot, this way we could use ML data to reduce road accidents. These unfortunate incidents actually form a pattern, and whenever we see a pattern, we can avoid it by using Artificial Intelligence models.
We finally hosted the web-app on heroku server, so as to provide connectivity worldwide. For the project, since we did not have any GPS sensor, we statically set our location to Sector 5, Salt Lake City, Kolkata.