Top-10 YouTube Channels for Data Science

Finding the best Data Science YouTube Channels by scraping the YouTube!

YouTube is one of the best resources where it involves all categories of videos across the world.YouTube has a vast amount of growth in recent years. In this blog, we are going to see a specific kind of category on YouTube that is the Data Science category.

In recent years Data Science is one of the best emerging technology around the world. So that there are many resources to learn Data Science online has come away, one of the best resources is that the YouTube channels for data science across the world. Now we are going to scrap the data from YouTube and make a visualization of it.

Overview of our work

Videos are a great learning tool, there are many youtube channels that are based on educational content across the world. Even though there are many channels, finding the best quality content channels are difficult. The channel that gives quality content will be liked by many peoples. So our idea is to target those channels that are with many followers(subscribers) are the best channels.

Web Scraping

YouTube does not provide data easily as other social media. The only way to get the data from YouTube is to web scrap the contents in YouTube. Scraping using python is a bit complicated way, so we are going to use the Octoparse tool to scrap the data. To list the channels that are based on Data Science we have to go to YouTube search and enter “Data Science” and also we have to do some filtering for channels. So in the filter option do the following steps,

1. Search Type: Channel
2. Sort By: ViewCount

Once you search and filter, copy the URL link and paste it into the Octoparse tool. The tool will automatically scrap the data from the given URL and give you the scraped data.

This data was scraped on 25-May-2021

Preprocessing the data in python

Now we have the scraped data, our next approach is to preprocess the data. No data is proper until then it is cleaned, in this problem also we have to do some cleaning process. The cleaning process involves,

• Removing unwanted columns,
• Renaming columns,
• Change the greek symbols to numbers.

By completing the above steps, download the cleaned data into your local system.

Python Code: https://github.com/selvankumar-m/WebScraping

Tableau Visualization

Now we are in the final stage, that is the visualization part. Open Tableau and connect to the data source.

As we are focusing on the top 10 channels on the basis of the subscribers, we have to create a new set in Tableau that takes the top ten values from the subscribers. And finally, we plot the graph of Top-10 YouTube channels for Data Science.

Tableau public: https://public.tableau.com/app/profile/selvan.kumar.m/viz/Top10YoutubeChannelsforDataScience/Sheet1

The Top-10 Channels are,

1. Edureka!
2. Simplilearn
3. Joma Tech
4. Intellipaat
5. Siraj Raval
6. StatQuest with Josh Starmer
7. Great learning
8. Krish Naik
9. Codebasics
10. 365 Data Science

Among these 10 channels, 6 of the channels are organizations that have a team and consistently working for the data science contents, other 4 are individuals also having connections with the organization for making data science contents.

Hope you got nice content from reading this blog. Stay tuned for more.

until then Happy Analyzing!!!