DEV Community

Cover image for Comtonlizer


chintukarthi profile image Karthikeyan Dhanapal ・2 min read

Hello Everyone,

This is about the new project that I have been working on which is still in beta and I thought of sharing this with the community to get the valuable feedback and improve it.

When a content creator uploads a video to YouTube, he/she usually goes through the comments section to get an idea of how people are reacting to their video. It's obvious that we can get info from the like or dislike button, but they can't really get to know about the viewer's emotions unless we skim through the comments section.

Sometimes we don't really want to read all the comments either because we have a lot of work and no time or it's just we need something to analyze it. While I was looking at a video, I really want to know how other viewers have expressed their emotions for one particular video. That's how Comtonlizer has born.

Comtonlizer = Comment + Tone + Analyzer.

I have used YouTube's API to get the comment for a particular video and using those comments as an input to IBM Watson's tone analyzer API to get the tone score and the tone score is used to display the results.

I have hosted the application in netlify for which I've given the link at the end.

Here is the landing page of the application.

Alt Text

The input should be a valid YouTube URL and only when it's valid, the user is allowed to click the analyze button. The output of the popup screenshot is attached below.

Alt Text

Here we can see that the output is expressed in terms of emoji and percentage.

  1. 😍 - Represents Joy percentage of comments. How much the user has enjoyed the video based on the comment given is analyzed here.

  2. πŸ™„ - Represents the ok-ish percentage of comments. This means that the comments are neither good nor bad but in a neutral zone.

  3. 😭 - Represents the sadness in the viewer's opinion. How sad they are based on the comment given by the viewers is represented in this emoji.

For now, I am analyzing only recent comments with a limited count since for a particular video there might be thousands and thousands of comments. So this application can be used to analyze the video with recent comments. I have plans to update this in the future to an extension where we can get the analysis on the same page without going to another page.

Link to the application:

Please access the site using laptop or system. Even though it's accessible on mobile, the best experience is with a laptop or system.

If you have any suggestions or feedback to improve the application, please do let me know in the comments.

Discussion (0)

Editor guide