A Web of Interconnected Interests: What YouTube Can Tell Us

Your Reading Habits Online

The Internet is an essential medium that permeates nearly every aspect of modern life. People can establish communities without geographic restrictions and access content about hobbies, the news, and as much knowledge as they can possibly intake.

But you already know all this. You’re reading this online — and the fact that you’re reading this probably means you’re interested in technology or data science. Your decision to read this, along with a multitude of other things you do on the internet create a reflection of the real you… just on the internet.

When this idea is scaled up and the proverbial dots are connected, we enter the realm of social network analysis (SNA).

Social Network Analysis

SNA is a process of mapping and measuring the flow of relationships and relationship changes between knowledge-possessing entities. Simple and complex entities include websites, computers, animals, individuals, groups, organizations, and nations. A graphical display of a social network is composed of two primary elements: nodes (individual players/actors in the network) and edges (relationship or interaction between the nodes). Through qualitative and quantitative analysis of a social network, researchers are able to gain valuable insights into the roles that different actors play within the network, as well as the nuanced connections among these actors.

An Interconnected Community Online: Analyzing YouTube

While social network analyses are often conducted on “important” topics — terrorist networks, the spread of misinformation, political networks, etc, SNA can also highlight the more mundane, to equally stunning effect. In our study, we leveraged network graphs to explore the connections between different YouTube videos based on their common audience. The larger the size of the common audience, the stronger the association between the two video themes. In the graphical display above, YouTube video topics by color, videos are represented by nodes (dots), and overlapping commenters are represented as edges.

The network is incredibly interconnected. However, some video themes are more centralized than others in the network (shown through their position in the graph and number of connecting edges). Stress was a particularly interesting topic given the pandemic. When testing the clustering coefficients of some of the larger clusters, we found that Sports videos shared the highest clustering coefficient — the strength of association between any two given topics — with videos about Stress.

Network Analysis Outside Lofi and Stress

Besides tracking the interests of online communities, SNA has other powerful applications in multiple areas, including disease transmission, business networks, information circulation, and criminal investigation. In many cases, the same principles still apply across domains. For example, an organization that uses social media to spread misinformation could be visualized where each node represents an individual piece of content, while edges can represent individuals creating or resharing this content. In fact, this exact network could be developed using video content regarding the 2020 US election.

Going Forward

Social network analysis is a powerful tool for revealing insights into our increasingly connected world. The beautiful data visualization and precise graph metrics not only shed light on mundane interests but have significant potential to improve social priorities. Properly leveraging social network analysis makes intricate social structures and relationships crystal clear, which can open up new possibilities for a wide variety of fields.

Works Cited

What is Social Network Analysis (SNA)? — Definition from Techopedia. (2012, November 3). Retrieved October 25, 2020, from https://www.techopedia.com/definition/3205/social-network-analysis-sna




Exploring the applications of AI in consumer analytics.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

21 CSCW 2021 papers to read

KASPR Datahaus Joins the Big Data Protocol Data Alliance Adding Global ICT Activity & Quality Data

OWL data analisys Week 4 Stage 2

Decision-drivers in the recruiting process

Vehicle Accident Severity Prediction Based on Environmental Stimuli Using Machine Learning

Data for Society: What, How, and Why

Frédéric-Back park, Montréal

Introduction to Pandas

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Basil Labs

Basil Labs

Exploring the applications of AI in consumer analytics.

More from Medium

What Type Of Tile Is Best For Your Home?

My Love for My Houseplants

Three skills that make you a more popular exchange student

My Favourite Podcasts To This Day