This package contains fake-news profiling models, as well as experimentation and evaluation tools.
Build with python setup.py sdist bdist_wheel.
Import into a local virtual environment with python setup.py install.
evaluation contains functions used for evaluating the fake-news classification models.
experiments contains Experiment classes used to test classification models as well as perform hyperparameter tuning
for models. The ExperimentHandler coordinates saving config data, loading datasets, executing experiments and
reviewing results.
models contains baseline and final fake-news classification models.
processing contains tools for preprocessing text data as well as extracting readability, sentiment, or named-entity
information from text data.