Child Safety Youtube Extension

• 

  What the extension looks like.

• 

  Filtered comments.

Who are we?

Hi, today our project focuses on detecting "hate" in Youtube comments. With a machine learning model imported via HuggingFace, we will censor comments on the Youtube website.

Inspiration

We've decided to pursue this idea because we've noticed a lack of Google-extensions that would help parents assist their children. When we look up Youtube child protection, we've noticed a lack of customization in Youtube's child protection app past pre-teen age of 13-years old to a 17-year-old.

What it does

This will rate how NSFW each comment in the comment section is. Based on this, we will censor through the data and will act based on settings we've created. These functions include: "child-safety lock", "Toxicity Percentage", "Custom filter words".

How we built it

We've built this project through a multitude of code sectios. First off, we start with the web-scrapping aspect. Then, the data is sent into the AI Model imported through the use of Google's firebase, a cloud server that stores data. The AI Model that will be used to reformat the data and to take in the input of "negative" comments. One of the function we created used an open source website on link

Challenges we ran into

We've run into a major issue of communicating the code between Javascript and Python. This has created a severe time-delay. Learning Javascript and creating UI/UX.

Accomplishments that we're proud of

• Creating a google extension
• Manipulating Data based on Scrapped Data
• Webscrapping

What we learned

• How to communicate between Javascript and Python
• Utilizing an open-source AI model.
• Utilization of firebase
• Javascript/CSS/HTML
• Basic DOM Mutation
• Chrome Extension creation

Built With

• chrome
• javascript
• json
• pipeline
• python
• render
• requests
• urlib