An AI / ML project idea to help reduce crime against bicycles in Stockholm
Final project for the Building AI course
People hate getting their bikes stolen. A website or app is set up for people to enter data about a bike theft in the city (location, day of week, time of day, bike value, bike condition, etc.). This data is processed to make an updating map of the city for theft hotspots, and eventually predict the level of security one should apply to their bike, and if they have a high chance of getting their bike stolen in a certain area.
Which problems does your idea solve? How common or frequent is this problem? What is your personal motivation? Why is this topic important or interesting?
- problem 1: We are living in a bike theft epidemic.
It is very annoying and costly. I have had 3 bikes stolen while living in Stockholm and I want to help prevent it from happening with the use of AI.
First, a victim goes to a website or app - basically a place for people to log and submit key details about a bike theft event the data is stored and processed the data is used to highlight high bike theft areas via an interactive map which helps to predict the likelihood of a bike theft recommends alternative parking areas or suggests the level of security needed to feel safe about your type of bike being left in that area based on brand, condition, style etc. The main goal is to prevent bike theft Additionally it could act as a historical bike theft event log The users are the people who have had their bike stolen in Stockholm. This could be expanded to other cities.
More opportunities to build upon this idea...
Images will make your README look nice!
Once you upload an image to your repository, you can link link to it like this (replace the URL with file path, if you've uploaded an image to Github.)
Data is collected via an app or website. Clustering, regression
What does your project not solve? Which limitations and ethical considerations should be taken into account when deploying a solution like this?
Would be great to work with an artist to make the map look good, ML engineer to help coding etc. More work on this soon.