Landing page

About Us

Exploring YouTube Video Popularity through Data Science

Project Summary

The aim of this project is to analyze the popularity of YouTube’s music video with Youtube’s mv features, combined with the corresponding Spotify’s artist popularity and the corresponding track popularity, also explore these features and YouTube music videos’ popularity through regression analysis, binary and multiclassification analyses, unsupervised learning, interaction rate, etc.

Research Questions

  • Is there a relationship between video features (such as mean sentiment score，comment count, subscriber count, and publishing time) and YouTube video views?
  • How can we predict Spotify’s track popularity based on singer popularity and YouTube video views or other related features?
  • Can sentiment analysis (sentiment score range: -0.5 to 0.2) help evaluate the interaction level between the video and its audience?
  • How can we predict the Like/View ratio (engagement rate) based on video features such as comment count, duration_seconds, etc.?
  • How do video features like mean sentiment score impact the likelihood of a video receiving high views?

Literature

Literature Review

In recent years, the rapid expansion of video-sharing platforms like Youtube has triggered a wide range of research aimed at understanding the factors that influence video popularity, viewer engagement, and content analysis. Lots of studies have explored different aspects of YouTube dynamics, such as the discovery of persistent tags and trending videos, to better understand user behavior.

Dokuz (2024) proposed a method for identifying popular and persistent tags from a dataset of trending videos on YouTube, revealing trends in the user-video relationship and how video tags influence YouTube comments for engagement and visibility analysis, which is also a prominent area of research as it provides valuable insights into viewers’ reactions and opinions about video content.Khin Nyunt and Naw Thiri Khin (2024) investigated how the combination of sentiment analysis and trend analysis can guide aspiring youtube users to choose a successful career path.Similaet al. (2024) focused on sentiment analysis techniques applied to YouTube comments to provide valuable insights into users’ opinions on videos. Their study highlights how sentiment analysis can be used to predict video performance, as positive sentiment is often associated with higher engagement.

Additionally, artificial intelligence and deep learning are becoming more prevalent in analyzing YouTube comments.Meghana (2024) provides a comprehensive overview of sentiment analysis techniques, evaluating how advances in Natural Language Processing (NLP) can improve the accuracy of sentiment categorization and address challenges such as multi-lingual comments and context sensitivity. These studies help to understand how content features and user responses interact to shape the success of YouTube videos.

References

  1. Dokuz, Y. (2024). Discovering popular and persistent tags from YouTube trending video big dataset. Multimedia Tools and Applications, 83, 10779–10797.
  2. Khin Nyunt, N. T., & Khin, T. (2024). YouTube Career Analysis with the Combination of Trending Analysis and Sentiments Analysis. ESS Open Archive.
  3. Meghana, K. (2024). Artificial Intelligence and Sentiment Analysis in YouTube Comments: A Comprehensive Overview. 2nd International Conference on Intelligent Data Communication Technologies and Internet of Things (IDCIoT), 1565-1572.
  4. Giri, R., Sirsath, M., & Kanakia, H. T. (2024). YouTube Comments Sentiment Analysis. IEEE 9th International Conference for Convergence in Technology (I2CT), Pune, India, 1-4.
  5. Dasovich-Wilson, J. N., Thompson, M., & Saarikallio, S. (2024). The characteristics of music video experiences and their relationship to future listening outcomes. Psychology of Music, 0(0).
  6. Efe, I. E., Tesch, C., & Subedi, P. (2024). YouTube as a source of patient information on awake craniotomy: Analysis of content quality and user engagement. World Neurosurgery: X, 21, 100249.

Additional Ideas for things to include

  • Audience: For data professionals, business analysts, and digital marketing strategists.
  • Headline: Unlocking the Power of Cross-Platform Synergy
  • Introduction: We offer insights into how data-driven strategies can enhance content visibility and viewer engagement on YouTube and Spotify.
  • Motivation: EThis analysis is crucial for content creators and marketers aiming to optimize their digital strategy in a highly competitive space.
  • Key Topics: Machine learning, Youtube & Spotify features, and data visualization
  • Call to Action: We can help music industry practitioners better understand and grasp the market. Music producers, companies and singer teams can analyze the correlation between YouTube and Spotify platforms to allocate marketing budgets more rationally and produce more targeted content based on the characteristics of different platforms.