Skip to main content

Twitter – Changing the Graph of Social Networking

Unlike the social networking sites preceding it, the open and asymmetric nature of Twitter has created a unique social experience. Before the emergence of Twitter, relationships in social networks were largely bilateral. In other words, if someone requested to be your friend, you would be their friend as well. Twitter handles relationship requests asymmetrically. Both parties must request each other before a bilateral relationship is formed. This design has diversified the social graphs for Twitter users. Most users of Twitter will fall into 1 of three categories:

The Follower – The person who will follow may other users of Twitter to remain up to date on their “news” (any Tweets). The tend to Tweet infrequently and have few followers.

The A-lister – The user that has many followers who he/she mostly does not know. This user will usually follow few people and most will be other A-Listers. More often than not, this user will Tweet frequently.

The Symmetric user – This user will follow approximately the same number of users that follow them. This user tends to use Twitter to stay updated on “news” within his close social network. Symmetric users typically do not have a social network exceeding 150 nodes.

The social network graphs formed by users can vary greatly. The diverse classes of users often have network graphs including many directed paths. The A-lister and the follower tend to have significantly larger networks than that of the symmetric user. The simplicity and openness of these large social networks has garnered the attention of marketers and has propelled Twitter to be on the forefront of real-time “news” delivery. However, despite all this, most users of Twitter have a very small group of “friends” (users that exchange Tweets to each other). After analyzing the large directed graph of many Twitter users it becomes evident that the underlying “friend” graph remains sparse and bilateral; similar to that of most other social networks. This is the true graph that matters most when investigating the transmission of messages around Twitter.

The underlying “friend” graph for users can help us accurately predict who they will transmit their messages to. Given user with both a Facebook and Twitter, it is expected that their “friend” graph would be similar on both networks. (Assuming most of their friends have both a Facebook and Twitter) The simplicity and openness of Twitter have allowed it to flourish as a news and advertising outlet. Two users with social graphs that are not connected can easily find each other by means of discovering trending hash tags. Likewise, two nodes that are not directly connected in a social network can become connected when a mutual friend “retweets” another node’s tweet. This open model for the social network allows users to share their “news” for the entire Twitter world to see. An example of the grandeur of the Twitter network was exemplified during the breaking of the news of “The Miracle on the Hudson.” A lot of the preliminary news information was broadcast by people on Twitter who were on the scene quicker than news reporters. (

Twitter provided users with the unique experience to create a very diverse social network. Despite the directed social graphs created within Twitter, the underlying “friend” graph remains similar to that of other bilateral social networking sites such as Facebook. None the less, marketers have seen these large networks as an opportunity for them to transfer “news” to its followers in hope that the followers re-transmit it to their social networks. The diverse graph structure enabled by Twitter has positioned it a social hub for friends, a marketing outlet, and a way for any person read and share “news” to the world.

Sources :

Twitter Networks are Different than Social Networks

Other Useful Information:


Leave a Reply

Blogging Calendar

September 2011
« Aug   Oct »