Social Network Subdivision Within Twitter
Link: https://journals.sagepub.com/doi/full/10.1177/2056305117691545
As social media has gained much prevalence in our society today, applications like Twitter, Facebook and Instagram have attracted people of all genders, ages and races. While all of these platforms have transformed into focal points of the global economy for ideas and opinions, none have reached the impact of Twitter. In Classifying Twitter Topic-Networks Using Social Network Analysis by Itai Himelboim et. al, the various ways in which connections exist on Twitter are outlined to show the versatility and variability of the platform. Twitter has the power to rally tens of thousands of people against one, kindle acts of political violence, create communities that are endless and strong, yet also momentary and breakable. And while all of those things are possible to an extent on other platforms, Twitter offers the possibility of relative anonymity. Facebook is defined by who your friends are: how many do you have? Where are they from? Who are your friends friends with? All of this is done with tons of transparency, there is a face and a community directly associated with an opinion. On Instagram, the main mechanism is likes. How much quantitative support did an image/video/caption receive? Twitter, however, is powered by comments. The comment section of a tweet is one of the most opinionated places on the internet. And the way the comment sections have gotten that way is explained in this study. The research breaks up Twitter’s patterns of information flow into six categories: Broadcast, Support, Polarized Clusters, In-Group, Brand, and Clustered Community.
Through the course of this class, we have explored the idea of numerous types of networks. In terms of things like social media, a network is a group or system of interconnected people or things. On social media, we are able to see what our friends are thinking, for social media has morphed into a platform that is widely used to share thoughts and beliefs. Twitter, as mentioned before, has become powered and driven by this aspect. It is a network not of individuals, but of ideas and opinions. The conversational comments that have blossomed from this area of Twitter has allowed for certain “groupings” to come about. These groups have been characterized as polarized clusters, which are spaces in which ideas bind a small group together to some degree of thought isolation. These are examples of homophily– the idea that people with similar ideas will seek most to interact with one another. There are various ways in which Twitter has followed this idea of people with similar interests virtually gravitating towards each other. An example of this happening via Twitter is through the idea that there are different political groups that have formed through the similarities in beliefs and ideas. In the world of politics, there have been social networks of predominantly left leaning users who have found themselves forming relationships and bonds through their political beliefs. On the other hand, there are also social networks with predominantly right leaning users who have also bonded over these similarities. These networks have been fostered by the ability that users have to communicate their beliefs and values to one another. Himelboim’s article describes quantification of these subnetworks, one of which is Broadcast groups, which have high centralization, meaning that the information comes from a small number of sources. News outlets like CNN or pundits from any subject like sports, music, or politics all belong to the Broadcast subset. It can be argued that these outlets are at the foundation of the five other subnetworks. They all rely on whatever information they receive from the Broadcast group, and the way they divide up and interpret that information or opinion breeds cluster communities, and in-groups.
These smaller more niche networks have come to fruition through this aspect of Twitter that we have not seen to this extent before. As we learned in class, the connectedness between users on social media is more intertwined than we imagine. Within this massive link between users lies these divided subnetworks of people. These divisions of networks provide the larger platform with some subdivisions. These subdivisions somewhat resemble some negative relationship examples that we have seen in class, where every member of one group dislikes every member of another group. This combativeness is especially highlighted on Twitter because of the multitude of subjects up for discussion. Many of these subdivisions also overlap with each other, allowing the rhetoric of one idea to flow into a completely unrelated topic. This overlap of polarized clusters has garnered some strange results on Twitter, such as a subset of K-Pop fans selling out a Donald Trump rally just to leave a large portion of seats empty, or the evolving voice of politics amongst conversations about sports and music.