By using 2020 March to August Yellow Cabs transactions try to forecast 2020 September daily and weekly transactions. While doing the forecasting try to add Covid, Seasonality,and any other external variables to the equation.
The solution can be found in notebooks/transaction_forecasting.ipynb
Marketing is trying to identify pickup location ids to incentivise passengers in order to increase the number of trips. By using 2020 March to August Yellow Cabs transactions build a recommendation engine to help marketing identify these potential location ids on a weekly basis? Also define why you think these areas are potential areas to incentivise according to you?
The solution can be found in notebooks/pu_location_recommender.ipynb
We want to promote tipping to retain our driver base and increase the driver engagement. Could you find the relation between [the trip duration, trips distance, location id, vendor and passenger count] and tip amount, and make suggestions on rolling out some experimentations?
The solution can be found in notebooks/increase_tipping.ipynb
Some visualizations and data clensing can be found in the notebooks/clean_trip_dataset.ipynb .
The sources, transformations and exploration can be
found in notebooks/external_data.ipynb
To make sure all the needed dependencies are installed run the following command:
conda env create -f environment.yml
To add the environment as a kernel to your jupyter notebook run:
python -m ipykernel install --user --name=free-now