Inspiration

Most youth spend countless hours on social media and encounter a lot of negativity and toxicity, which in turn negatively impacts their mental health. For example, Youtube is very popular social media platform but also is frequently bombarded with negative comments. We want to provide a way for individuals who still want to engage in social content without being negatively triggered.

What it does

Our solution aims to keep the original intention of social media as a positive space for people to interact by converting negative comments to positive ones. Our Google Chrome Extension detects comments that contain aggression or negativity and replaces with a happier comment using DOM.

How I built it

Using Google's Natural Language API, it communicates with the Chrome Extension using a Rest API. The Chrome extension scrapes youtube comments out of the window and sends that information to the Rest API using an HTTP post request. Our Rest API then uses the analyze_sentiment method from the Natural Language API to determine if it is a good or bad comment and returns the answer to the client. Based on the result, the client replaces the negative comment with a positive one.

Challenges I ran into

How to set up the Google Natural Language API and its service account credentials. We also ran into a CORB error which we could not fix completely and handled it by disabling security features on Chrome.

Accomplishments that I'm proud of

Integrating the natural language API using chrome extensions.

What I learned

The interaction of APIs and chrome extensions. We learned how to use the google cloud platform including service account credentials, Natural Language API methods, and modifying HTML pages using chrome extensions.

What's next for PosCom: The Happy Comment Filter

Incorporating PosCom on more platforms like Facebook, Twitter, etc. The future iteration of our extension will fix the CORB error.

Built With

Share this project:

Updates