This project is designed to parse in all the tweets from a specific user and use NLP to generate a new, fake tweet. The tweet will then be displayed alongside a real tweet on a web page, and the user will be prompted to choose which one they believe is real. Once they make a selection, the correct answer and the incorrect answer are appropriately highlighted. Stretch goals include: tweeting out those tweets which succeed in stumping a user, adding some level of machine deep learning, and caching stumpers IOT tweet out only the highest success rate stumpers.
Authors:
- James Feore ([email protected])
- David Lim
- Miguel Pena
- Elyanil Castro
June 4, 2017 Repository created and initial frameworking completed.
- TBD
- Clone this repository to your local machine.
- Change directory into "turingtweets"
- "pip install -e ."
- "python -m nltk.downloader all"
$ cd turingtweets/
$ pip install -e .
$ pip install -e .[testing]
$ tox