Awesome

Simpsons Markov chain Twitter bot

A Markov chain model to generate text and tweet it. In this particular example we used ~600 chapters of the Simpsons to generate the Homer Simpson text model.

Bots

Homer Simpson's bot

Getting Started

In the the notebook explore_dataset_and_create_bot.ipynb you will find the code snippets to generate a Markcov chain generator model trained using a dataset. For this particular model we choosed to train a model based on Homer Simpson.

Prerequisites

See requirements.txt

Running the bot

Follow the steps on explore_dataset_and_create_bot.ipynb
Setup a twitter application
setup keys.py
run python bot.py model_name. Use the same model_name that you defined in the notebook

Setup keys

Generate a the file keys.py that contains

API_KEY = 'FILL_API_KEY'
API_SECRET = 'FILL_API_SECRET'

ACCES_TOKEN = 'FILL_ACCES_TOKEN'
ACCES_TOKEN_SECRET = 'FILL_ACCES_TOKEN_SECRET'

and save it in ./

You will get this keys following the [tutorial]https://www.mattcrampton.com/blog/step_by_step_tutorial_to_post_to_twitter_using_python/

Built With

markovify - A simple, extensible Markov chain generator.

Authors

Ivan Lengyel http://ivanlen.github.io/

See also the list of contributors who participated in this project.

https://github.com/jsvine/markovify

This project is licensed under the MIT License - see the LICENSE.md file for details