Information Cascades in Social Media: Crisis and Viral Trends
Articles:
Information Cascades in Social Media in Response to a Crisis: a Preliminary Model and a Case Study
Predicting Successful Memes using Network and Community Structure
Seeking a respite from his immense workload, Harambe logs into Facebook. Boom! A couple of hundred memes pop up on his timeline. Harambe is irritated. “These memes make absolutely no sense at all! Why do people keep sharing them!”, he exclaims. Harambe has spent hours at a stretch editing a picture he took the other day. He also put it up as a profile picture and this was his ‘best picture yet’ according to Harambe. Do you know how many likes he received? None! Harambe wondered how these memes became so popular. Like a mechanical arm stuck in a non-creative limbo, Harambe desperately tries to reach the end of the seemingly endless cascade of memes. Wait, wait, wait… Something caught Harambe’s eyes. It was a picture of a cat with an unusual expression on its face. The caption read, “The Grumpy Cat”. Like all the posts preceding this, it made no sense to Harambe, but for some reason he could not stop laughing! Driven by a hypnotic desire, Harambe clicked on the share button and bam! The post was liked, commented on and shared by a million other people in an instant. A plethora of emotions now flooded through Harambe. Should he be elated that people loved a post he shared, or should it bother him that picture of a cat got more appreciation than his profile picture. He decided to settle for the first one and that, my friends, is how The Grumpy Cat went viral!
One who is not familiar with the countless memes centered on Harambe will probably not be able to connect with the paragraph above. In that case, I will ask you to go online and look for these memes and try to figure out why I tried to explore, through the paragraph above, how Harambe would feel about these information cascades if he were a human being. Information cascades is an important feature of any networks as shown in our course structure. At first glance, it may seem that it is not possible to predict cascades of information. It may also appear that these memes are products of the whims of social media, but in reality, there are a number of factors which decide the cascades of information. In a study conducted at Rutgers University, researchers studied the flow of information in Twitter during an emergency. Tracking the number of retweets following an emergency broadcast, they were able to plot a graph of tweets after the broadcast. In their graph, there was a directed edge from User A to User B, if User A retweeted a message by User B. Most messages terminated after they have been shared once, resulting in very few long chains of retweets in the data. It was found that these cascades originate from local media users and tend to be wide, reaching a wide variety of users in breadth but not very long, reaching a narrow variety of users in depth.
On the other hand, in case of a viral outbreak, the scenario is quite different. A group of researchers demonstrated that the popularity of memes depends on three factors: positions of early adopters in network, community diversity, early growth of a meme which predicts future popularity. It was demonstrated that viral memes tend to have a higher early growth rate than non-viral memes. They were shared continuously and did not terminate after they have been shared once. It was also seen that viral information cascades originate from early adopters and tend to be both wide and long, reading a wide variety of users in breadth and depth. However, what caused these to have higher early growth rates than other internet phenomenon is still a mystery, isn’t it?
So, how and why is it that the internet is flooding with memes centered on a creature who was killed, and that too for no apparent reason?