Takes in a text sample; if the sample is at least 100 words in length provides a profile of the author based on writing style.
Uses the bag-of-words model with TF/IDF term weighting.
Currently returns author gender with approximately 70% accuracy.
Basic app is deployed here: http://ec2-54-68-86-232.us-west-2.compute.amazonaws.com/
To run the app yourself, clone the repo, run the command pip install -r requirements.txt
, and then run views.py.
Feel free to suggest additional features for analysis!