Print Page   |   Contact Us   |   Sign In   |   Register
Complexity Matters
Blog Home All Blogs
Search all posts for:   

 

View all (416) posts »
 

Different Network Structures in Twitter Maps

Posted By Prucia Buscell, Thursday, March 27, 2014

Twitter conversations create identifiable networks that have structural differences depending on the topic and the influence of dominant individuals. The structures are created as participants in the network choose the people they answer, retweet, and mention in their own messages, according to the Pew Research Internet Project.

The Pew researchers found six identifiable network structures: divided, unified, fragmented, clustered, and inward and outward hub and spoke structures. The report summary contains explanations and examples of teach type.

People who tweet about political topics, for instance, tend to form divided structures, in the form of two separate and polarized conversational networks. Participants in these differing networks don't interact with each other and they rely on very different sources of information. In many controversial conversations participants in the networks that Pew identified as liberal or conservative cited very different websites and distinctly different words and hashtags. People in the liberal groups generally cited URLs for mainstream news outlets, whereas conservatives tended to cite URLs for conservative news and commentary websites, the report says. The report says the finding underscores the partisan nature of political tweeting and group reliance on different people and organizations as well as different news sources. It also shows the two groups usually ignore each other despite intense interest in the same topics.

Unified networks are tight crowds of highly interconnected people often joined together by professional interests and hobbies. These structures often show how networked learning communities work and how social media can foster sharing and mutual support. People who form Twitter groups based on their interests in brands, products or celebrities, tend to form fragmented networks because they focus on their interest, but don't usually connect with each other.

Clustered conversations often look like bazaars with many centers of activity, the report says. For instance, people interested in the disappearance of Malaysian Airlines flight 307 could follow the news presented in several languages by several news outlets. Any global story, the report says, can generate multiple and diverse audiences that illustrate diverse opinions and perspectives.

Broadcast networks tend to form a hub and inward spoke structure, in which participants repeat and comment on the output of well known media outlets. Participants are often connected to and in conversation with the hub, not each other. Support networks, such asbusinesses trying to resolve customer complaints, create a hub and outward spoke structure, where the hub business sends replies and information to many disconnected users.

Social media is the new public square, Pew researchers say, and the network maps formed by Twitter conversations are like aerial photographs that show size, composition, and network locations that are analogous to positions of strategic importance in physical landscapes. These locations can help identify key people who influence social media conversations. Read the Pew Research Internet Project report for more information, illustrations of the maps, and further sources on network data and visualizations.

Thanks to Buck Lawrimore for pointing out this story.

Tags:  buscell  complexity matters  connection  systems 

Share |
Permalink | Comments (0)
 
Association Management Software Powered by YourMembership.com®  ::  Legal