Protekt

Protekt: Montreal's wingman against crime. We map danger zones, spill the tea on crime hotspots, and slide into your notifications. Because who needs surprises? Get it now and stay street smart!

Comment

Inspiration

We knew that becoming batman was not physically possible. Instead of waiting for him we built it ourselves.

What it does

Montreal's safety at your fingertips. Our app analyzes crime data, maps danger zones, and sends real-time alerts.

How we built it

Our full-stack marvel combines Python, Node, Flask, and React, with the mighty Google Maps API. Leveraging Montreal's open data, we dive deep into crime statistics—crime types, borough-wise incidents, and population metrics.

Backend Wizardry:

We wield Python, Node, and Flask to orchestrate a symphony of data. With Montreal's open data as our muse, we calculate a weighted mean, assigning crime types different weights for a bespoke danger score. MongoDB, our trusty vault, stores this trove of information, ensuring swift data retrieval for a seamless user experience. Our custom Flask API unveils endpoints, orchestrating a seamless symphony behind the scenes.

Frontend Mastery:

Behold the visual feast on React-powered Google Maps! We drape the island of Montreal in heat circles, each pulsating with crime data. Clickable circles reveal a dynamic realm of information—scores, crime insights, and tailored statistics for each neighborhood.

Safety in Clicks:

Tap into a neighborhood, unearthing a treasure trove of safety intel. Our app transforms complex data into a dynamic summary, presenting danger scores and crime insights at your fingertips. Every click is a step toward informed decisions and a safer Montreal.

Challenges we ran into

  • Creating an aggregation framework based on multi polygon geospatial algorithm.
  • Issue with the deploying of the servers
  • Reconciling multiple datasets into a coherent output (UTF format)

Accomplishments that we're proud of

  • Completing the project entirely
  • Battling with google apis
  • Figuring out a relevant score metric
  • Sleeping 20 minutes ## What we learned
  • MongoDb as a cloud database
  • Manipulating hundred of thousands of data
  • Rendering dynamic visualization with React
  • Implementing a relevant score metrics using applied statistics ## What's next for Protekt
  • Dynamic map based on time series data
  • Real time user feedback
  • Geolocation notification
Share this project:

Updates