Pollen
is a project that aims at finding and understanding how pollen concentration is affected by weather. The
main idea stands in the usage of Bayesian Networks that effectively explain out the relationships in the model. The
dataset used is built thanks to information gathered from different sources such as
POLLNET
and ilMeteo for the city of Florence, Italy. In particular the
following attributes are taken into consideration:
- Wind
- Wind Direction
- Temperature
- Pollen Concentration
- Precipitations
Check more in the Report HERE.
To get a local copy up and running follow these simple steps.
The project provides a Pipfile
file that can be managed with pipenv.
pipenv
installation is strongly encouraged in order to avoid dependency/reproducibility problems.
- pipenv
pip install pipenv
- Clone the repo
git clone https://github.com/CrisSherban/Pollen
- Enter in the project directory and install Python dependencies
cd Pollen
pipenv install
Here's a brief description of the files and a directory tree:
* datasets (directory for datasets ) * out (directory containing output graphs ) * resources (directory containing partial datasets and original files) * src * creates_dataset.py: creates dataset from files in resources directory. * get_pol.py: selenium scraper that downloads pollen concentration from Pollnet. * network.ipynb: main file with documented outputs.
Output Network:
We can appreciate that we can now understand from the model that mid/high temperature is linked with an incresed concentration of grass pollen.
FAIKR-3 © Course held by Professor Paolo Torroni