Forecasting the presence and intensity of hostility on Instagram using linguistic and social features
This notebook provides code to reproduce the primary figures and tables in the paper
Ping Liu, Joshua Guberman, Libby Hemphill, and Aron Culotta. "Forecasting the presence and intensity of hostility on Instagram using linguistic and social features." In Proceedings of the Twelfth International AAAI Conference on Web and Social Media (ICWSM'18)
Note that while all data used was publicly available, in order to respect user privacy and Instagram's terms of service, we are unfortunately unable to share publicly the raw data needed to replicate the results in this notebook.
- Replication.ipynb: Jupyter notebook to run main experiments.
- u.py: data analysis code used by Replication.ipynb
- *.pdf: figures written by Replication.ipynb
- requirements.txt: python library dependencies
To make sure you're using the same version of all dependencies, you can create a virtualenv and install all dependencies listed in requirements.txt prior to running the notebook.
cd icwsm-2018-hostilityEnter repository directory.
virtualenv icwsmCreate a new virtual environment in the directory
source icwsm/bin/activateActivate the environment.
pip3 install -r requirements.txt(or just
pip) Install all dependencies.
jupyter notebook Replication.ipynbStart notebook
When you're done running the notebook, you can deactivate the virtualenv and remove the virtualenv directory
rm -rf icwsm