(The repo is for the Final project of IS 590 Data Visualization at UIUC)
1 Introduction and Preparation
2 Trim part
3 Visualization Part
4 Conlusion
The above navigation table of contents shows the general framework and construction of my work. It is dividied into 4 parts: Introduction and Preparation part, Trim part, Visualization part and Conclusion part. In the Introduction and Preparation part, there are the introduction of the project, instrcutions to scan this notebook, some usual imports and check the dataset we need to use again. During the Trim part, I do some clean and edit to better prepare for the later visualization, specially here, I solve the two probelm: Unified time type and missing category name (only id). Then, the most important part of the whol project is visualization part, which shows interactive visualization to help users better discover the dataset. And the last conclusion part includes some conclusions and feelings I get throughout the proccess.
Since the foundation of Youtube online video in 2005, it has made a huge set of breakthroughs that surprises everyone. It is one of the world-famous video sharing website and maintains a list of the top trending videos on the platform. There are various kinds of videos on it and we can’t even possible to know actually about its trending unless we do the analysis from the data. So, this project is meant to explore the secrets behind different categories in US Youtube videos.
Here the data I use is called US videos, which is a daily record of the top trending YouTube US videos and was collected using the YouTube API, including the data of the video title, channel title, publish time, tags, views, likes and dislikes, description, and comment count from year 2006 to 2018.(Links are listed below)
https://en.wikipedia.org/wiki/YouTube
https://www.kaggle.com/datasnaek/youtube-new/kernels
https://fivethirtyeight.com/features/media-coverage-doesnt-actually-determine- public-opinion-on-the-economy/
https://fivethirtyeight.com/features/a-comic-strip-tour-of-the-wild-world-of- pandemic-modeling/?cid=referral_taboola_feed