While listening to Mo Bamba by Sheck Wes, we had an idea: What if there were a way to judge the response to a song based purely on its youtube comments? Then we went broader and asked: What if there were a program that graphically displayed how the sentiment of the youtube comments changed over time? And that's when we realized that Wolfram had just the tools we needed to do such a task. We then began our research on Wolfram and the Youtube API and haven't stopped working since.
What it does
Our program comes in two parts. First using Python, we gather all of the youtube comments and the dates of the comments from a given video, and output a text file with all of this information. Then using Wolfram we input the text file and output graphical displays of the data, including change in sentiment over time and overall comment sentiment.
How we built it
Our program was built in Wolfram and the Google Collaboratory, using Python.
Challenges we ran into
We had difficulty combining Python and Wolfram and the Youtube API was also very dense.
Accomplishments that we're proud of
Over the course of the past 24 hours, all of us have learned Wolfram, a language that none of us have ever used before.
What we learned
We learned how to use Wolfram, the Google Collaboratory and the Youtube API.
What's next for Youtube Comment Sentiment Analysis
Next we will create a website for our code and make the process more intuitive automatic.