EvalTube

A web app that collects the comments of a youtube video, processes them using Symbl.ai, and returns information about the positivity of the video.

Comment

Inspiration

The Internet's anonymity allows for the complete freedom for whoever to say whatever without consequences. Whether or not this is good or bad depends on what is being said. Nowadays, it often feels it has swung in the direction of bad, with most people using the invisibility as immunity towards repercussion. Nowhere is this more evident than in the YouTube comment section. You would be hard pressed to find a single video without comments insulting the creator, fellow viewers, or entire groups of people based on race or gender. This makes for an altogether unpleasant user experience.

What it does

EvalTube allows for users to quickly and easily check a video's level of toxicity. First, a user inputs a video link on EvalTube's website. This link is sent to the back end, where the comments of the video are acquired using Google's API. Then, the comments are processed and analyzed using Symbl.ai's sentiment analysis. Finally, an overall score and some other statistics are passed back to the website to be displayed.

How we built it

The front end was built with HTML, CSS, and JS. The back end was built using Python with Flask. The API's that were used were Symbl.ai and Google's Developer API.

Challenges we ran into

We were all new to using the 2 API's, so figuring out how to use those was a difficult process. Another challenge we had was connecting the front-end and back-end.

Accomplishments that we're proud of

Learning how to use the new technologies and programs was difficult, and being able to use them in such a short period of time for us was something we were definitely proud of.

What we learned

We learned how to use new technologies (Symbl.ai, Google API, Flask). We also learned how important it is to have a solid plan and foundation.

What's next for EvalTube

One next step could be to add a database to store processed videos and their information so that it's even more accessible for users. Another possibility would be to migrate to an extension to be able to flag and block videos on the actual YouTube website.

Built With

Share this project:

Updates