Different sources cover the same news in very different ways! This project scrapes a dataset of article titles, sources, political leanings, and text from curated articles on allsides.com.
The scrape_data notebook scrapes the data and saves it into a pandas dataframe (cached in the data folder).
The data_to_html notebook loads that dataframe into an html file, which can be viewed in this app.
This data could also be useful for all kinds of tasks, such as classifying bias of different titles as well as trying to learn to translate between the same title with different political leanings.
- uses data from allsides.com
- horizontal timeline from codyhouse
- made by @csinva_