Social Media, Election and Threshold cascading
http://www.nytimes.com/2016/11/14/technology/facebook-is-said-to-question-its-influence-in-election.html
After the results of the 2016 presidential election in the United States, a lot of people have been trying to understand the effect of social media in shaping the outcome of the election. The above author mentions “filter bubbles” among users who largely share the same opinions.
If we relate this to what we have seen in class, these “filter bubbles” can be viewed as clusters and the information shared can be seen as the new behavior that is to be adopted. People on facebook exist as cluster, having strong and weak ties with their friends. If we consider the two major clusters based on political belief, Liberal and Conservative, we see that while these clusters have a lot of strong ties within themselves, there aren’t too many strong ties between clusters. Most of the ties between these clusters are sparse and weak ties. If a news article is shared by a few people in cluster L, it will be shared widely, given that there are a lot of strong ties and so the threshold value does not need to be particularly low. However, a person from cluster C might not agree with the same news article and the high threshold makes accepting and sharing new information more difficult.
This is usually what happens on social media, and while news articles can become saturated within one cluster, they will might not even be known by nodes in other clusters. Hence, though the information might spread fast, it does not spread evenly across demographics.