Media is no longer being objective when they write articles online. They support their own political point of views and try to sway people's opinions. We are fed up with biased articles and the negative effect they can have on people. We wanted to build a tool to help users focus on the facts.

What it does

It analyzes the sentiment values of every sentence in the selected article and picks up adjectives, adverbs, comparative and superlative words with sentiments in sentences. Then, it replaces those words with synonyms or deletes them to make the sentences neutral. The extension will display the article after modifying and highlights changed words. Users can hover mouse upon the highlighted words to see the original words in the article.

How we built

We built a chrome extension to do web scraping/front end and then a python backend to do content analysis. The backend then sends the modifications to be made to the front end which alters the article to reflect the changes.


The challenges we faced were related to the JSON object handling (which contains the modifications to be made) and data analysis/manipulation. We were not getting accurate suggestions from the library for replacement as well.

Accomplishments that we're proud of

At the start we did not realize that the project would be so tough to implement and with novice members on the team it feels great to be able to finish a version of the idea we had in mind at the start of the event.

What we learned

Team members had to ramp up quickly on technologies like javascript, use of APIs, etc. We believe everyone has a greater understanding of natural language processing library usage and JSON object handling now.

What's next for Content-Fiddling

Next is to add grammar checking and add a slider to adjust the strictness of the content modification based on sentiment values.

Share this project: