Sentiment Analysis on YouTube Data: TextBlob vs. VADER - A Comparative Showdown

In today's digital age, social media has become an integral part of our lives, with YouTube being the largest video - sharing platform globally, generating massive user comment data daily. Hidden within this data are users' genuine feedback and emotional tendencies towards video content. Effectively mining and utilizing this data has become a focal point for video creators, platform operators, and marketing professionals alike. This article takes you into the world of sentiment analysis on YouTube data, revealing the performance showdown between the TextBlob and VADER tools.

Website screenshots

Haowen Zhang from the University of California, San Diego, conducted an interesting and practically significant study. He collected 18,000 comments from 60 videos across 6 categories and analyzed them using TextBlob and VADER sentiment analysis tools. The results were then compared to the videos' likes/views ratio to assess quality. The study found that TextBlob's scores were more stable and positively correlated with the likes/views model, indicating its potential superiority in video quality prediction.

In the era of information explosion, people increasingly rely on videos for information and entertainment. The removal of YouTube's public dislike counter on November 10, 2021, made it harder for users to judge video quality based solely on platform - provided data. Sentiment analysis technology emerged as a solution, offering a new perspective for video quality assessment by analyzing the emotional tendencies in comments.

This research not only provides video creators with a basis for optimizing content but also offers valuable insights for platform operators and marketing personnel. By understanding user feedback through sentiment analysis, more targeted strategies can be developed. As data volumes grow and analysis techniques advance, the application of sentiment analysis in video quality assessment will become increasingly widespread and in - depth.

#SentimentAnalysis #YouTubeData #VideoQualityAssessment #TextBlob #VADER #ComputerScienceAndTechnology

The study was published in Advances in Computer and Communication, Hill Publishing Group

https://www.hillpublisher.com/ArticleDetails/4437  

How to cite this paper:  

Haowen Zhang. (2025) Sentiment Analysis on YouTube Data: A Comparison of TextBlob and VADER. Advances in Computer and Communication, 6(1), 35-40.

DOI: http://dx.doi.org/10.26855/acc.2025.01.006