Youtube Video TrendingAnalysis Based on Machine Learning

Abstract

Social media platforms play an important role in commerce, entertainment, marketing, education, media and communication. YouTube has a large and active user base, making it the center of corporate digital marketing efforts. There are always videos that attract a lot of attention in a short period of time and become trending videos on YouTube, and these videos can be displayed as trending videos on YouTube and updated daily. In this article, we analyzed data on YouTube trending videos in the US region and analyzed the targets that captured the attention of users in a relatively short period of time. We conduct exploratory data analysis on each aspect to gain data insights and find statistical similarities between them to understand viewing patterns across video categories. We present our analysis by measuring, mining, analyzing, and doing ANOVA comparisons of four scales of different categories of likes, dislikes, opinions, and comments. In addition, the study compared the accuracy of three machine learning methods on YouTube data, which counts the number of views and viewer reactions of popular videos on YouTube.