The Majority Illusion
http://www.technologyreview.com/view/538866/the-social-network-illusion-that-tricks-your-mind/
This essay discuessed an illusion of social network. For example, something, like the message, picture or the idea, is rare actually, but some people think it is common based on the information and the influence from their social network. Some people from USC have worked on it and discovered the reason for this illusion. Think about the paradoxical nature of social networks. There is an obvious case. Some special people will skew the average in some field, like the height. Similiarly, if there is a local social network, with 14 nodes, the different distributions of colored nodes (with the same number) have the different effect of the remaining nodes. Obviously, colored nodes with more neighbors (friends) are more powerful than nodes with less neighbors. This can cause the majority illusion. The majority illusion explains many phenomena. For example, some content can spread globally while other simliar content does not. It doesn’t matter in some cases, like the memes on Reddit or videos on YouTube. While it is to blame in others case, like antisocial behavior.
As we learned from the class, the social network can really affect people’s decisions and opinions. Restaurant example explains the reason people are affected by others’ decisions. The information based and direct benefits. We also know that we often don’t care as much about the full population’s decisions as about the decisions made by friends and colleagues. People are affected more by friends than by the straighter. Thus, if a person has more friends, s/he can influent more other people. Think about the example below, A has 4 friends and all of B, C, D and E have only friend. This example is extreme, but can explain the majority illusion effectively. If A is colored, all of B, C, D and E are going to be colored. However, if any of B, C, D and E is colored, A has low possibility to be colored since it has four friends and only one is colored.
