The goal of this project was to understand the relationship between the sentiment of YouTube video comments and the number of "likes" a comment receives. Comments and each comment's number of "likes" were scraped from YouTube and comments were analyzed for sentiment. The results of the analysis showed that the more positively sentimented a comment was, the more "likes" a comment received.
The implication of these findings is that positivity increases engagement. If a platform were interested in increasing engagement, they may wish to build an algorithm that promotes positively sentimented content.
-
Python was used to access YouTube's API to scrape YouTube video comments and the number of "likes" a comment received. Python was also used to access IBM Watson's Natural Language Understanding tool, which analyzed each comment for sentiment. Both Python scripts (“Channel1 - 6.25.19.py” and “NLU_youtube_comments.ipynb”, respectively) are stored above.
-
SAS was used to fit the statistical model of interest, and the SAS script “CMN Project.sas” is stored above.
-
A Powerpoint file summarizes the project and is stored above.