Twitter network analysis: Developing and Identifying Meaningful Content

Identifying influencers and innovators :

A person's position in a network is an indicator of how influential they are.This is because influential people are more likely to acquire connections, but also because having more connections in itself makes one more influential. Put another way, it's difficult to imagine how someone could be influential on Twitter without being followed by others interested in their topic. Without followers, how would they be able to express their influence?And if they somehow managed to be influential anyway, wouldn't they then acquire followers?

Knowing this , we can use algorithms such as eigenvector centrality and Page-rank to rank members of a network, and find out who is most influential.These algorithms reward people with lots of followers or relationships, but they also reward people for relationships with others who are themselves well-connected.In other words, these algorithms allow influence to be passed on. They search out those who are not only well-connected, but are connected to others who are central.

A variant measure when we're looking for influencers is 'betweenness'. Rather than finding people who are at the centre of communities, algorithms like betweenness centrality find those who are on the most paths between others in the network.

People with high betweenness tend to be the innovators and brokers in any network. They combine different perspectives, transfer ideas between groups, and get power from their ability to make introductions and pull strings.

Detecting subgroups in the Twitter network

Making subgroups(based on content and influence) is a useful way to probe any community and work out how it divides up between related interests and concerns.

Network visualizations also encode attributes as position, so users close together in the visualization tend to be closer together in outlook, expertise and attitude (because they are connected directly and/or share mutual connections).

In most of the twitter networks, sub-groups overlap and don't have clear borders. A person can also be a member of several groups at the same time.

Subgroup clustering can be done using various methods depending on content such as connectivity, centroid or density based clustering .

Example of the Econsultancy Twitter network segmented into communities

Using this network insight to develop content

In offline networks, influence tends to map loosely (but not completely) onto rank and positions of authority. Online and in social media, however, influence is much more about platform-specific qualities, such as the quality of the content you share.

For this reason, looking at online influencers gives us important clues about the type of content we should create. The content an influencer has shared is often what has made them influential, giving them the reach into the community we aspire to.

By segmenting influencers by sub-group, we can also develop a content strategy that reaches all parts of the network with equal effectiveness.

For example, we may have planned to create content that would go down a storm with a certain group in the network, but leave others cold. We might be able to tweak it to appeal to all members of the network, or have different content strategies for different sub-groups.

Summary

1. Firstly, we look to develop relationships with the people we've identified as having high centrality or betweenness. These are the people with the most reach into the network, the most visibility to others, and the most power to make and break messages.
2. Next, we look to make sure we're working with a set of influencers who are able to reach the whole of the network. It's pointless working with a group of highly visible people if they are all highly visible to the same small group. We need to make sure we have a good spread and can reach all the people who matter.
3. Finally, we should optimize our content. The influencers and sub-groups we've identified give us clues about what our community finds interesting and what kinds of content they are looking for.