Inspiration
We find the data privacy problem to be very important in the world, so we decided to create a tool to raise awareness and clear misconceptions about data privacy.
What it does
It uses statistical data from 500 popular websites to calculate the average score and find websites with similar tracking techniques.
How we built it
We used privacyscanner to retrieve the data about those websites, and than that data trough UMAP dimension-reducing algorithm to build the map of websites.
Challenges we ran into
Finding the proper way to calculate the average score and creating pretty frontend for visualizing.
Accomplishments that we're proud of
We turned huge amounts of unclear data to simple and clear diagrams that are easy to understand. That is the thing we loved most about this project.
What we learned
It was the first time we employed machine learning algorithms on hackaton and it turned out rather good!
What's next for Privacy benchmark
Probably providing more detailed output and adding more text explaining the results in a simple way.
Github:
https://github.com/hack-chain/hackaburg-2019
Forked repo for data extraction
Built With
- flask
- javascript
- python
- sklearn
Log in or sign up for Devpost to join the conversation.