The inspiration was drawn from the prevailing crimes in our community. This situation has caused disbelief in security agents, allowing our community to come up with vigilante groups. The vigilante group at a point tend to take the laws in their hands, and do what they think is right. The solution to this problem is to come out with a crime awareness app for both the security agents and the people leaving in the community .
What it does
The app is designed to study and classify crime data based on locations, and rank them into three awareness. The awareness is ranked as low(1), medium(2) and high(3). The app is designed to target the security agents in order to help them prioritize security levels depending on our classifier. The app can also help increase crime awareness to the residents in the community. They should be able to take precautious measures in securing their properties.
How I built it
Jupyter notebook is built to read the data file on crime. The data were trimmed to get location of crime (x and y), population, assaults, murders, robberies, and rape. The data types were explored in order to choose features for classification in deep learning. Jupyter notebook is connected to ArcGIS pro to have a display on a map.
Challenges I ran into
The challenge the team faced was using pytorch deep learning framework for the first time. The document was not easily picked up.
Accomplishments that I'm proud of
We accomplished having the framework in our classification model for crime.
What I learned
we learned the power of pytorch in artificial intelligence research.
What's next for Maptel Engine
The team is looking at incorporating time in our classifier. This should be able to alert the residents in time based on their location.