What it does
How we built it
We first conducted EDA and found some problems in the original dataset. Then, we clean the data based on that. After that, we build our predictive machine learning model using random forest with a RMSE lower than 0.5. Also, we put our data into a database using python sqlite3, and then we only provided users with a query function
Challenges we ran into
Understanding the differential privacy theory, cleaning data, and applying the theory to our model.
Accomplishments that we're proud of
Data preprocessing, EDA, ML algorithm, application of DP theory, and the ppt.
What we learned
Differential privacy theory.
What's next for Differential Privacy Project based on yellow cab data
Improving our Differential Privacy Mechanism and providing more useful query API for users of our database.
Log in or sign up for Devpost to join the conversation.