sherirosalia / a.i._president

M.L. generated tweets based on web scraping twitter and Markov generated text using splinter, beautiful soup and ML.

A.I._president

Machine learning applied to generate tweets based on data scraped from Twitter.

Webpage

Results of the project are hosted in GitHub pages: A.I. President

Custom Script Tweets

Markovify Script Generated Tweets

Folder structure

• docs holds html, css, image, javascript and json files for display site
• python folder contains scraping, file conversion and markov chain scripts
• initial data from webscraping is in data folder

Frameworks, Languages

• Python
• Beautiful Soup
• Selenium
• Splinter
• Markovify
• Pandas
• Javascript
• Html
• CSS

Deployment

The core files are in the python folder which contains both Jupyter notebooks and stand alone scripts. Note that Twitter changes the site regularly so selectors will likely need to be updated for the webscrping code to work. File paths will also need to be adjusted.