What is social
network analysis?
As a motto said, it does not matter what you know, but
who you know.
Social network
analysis is the study of the pattern of
interaction between actors. Actors can be people, community, group,
organization, country and so on. The phenomena and data reflected from the
relationship models of these actors are focus of social network analysis. In the
view of social network, the interaction between actors can be presented as a
models or rules based on relationship, and the regular pattern of the relationship
reflect the social structure, so the structure of the quantitative analysis is
the start point of social network analysis.
Therefore, social network analysis focuses on the
social actor and the relationship between the actors, as Fig 1 shows. The nodes
in the network are the people and groups while the links show relationships or
flows between the nodes. Social network analysis provides both a visual and a
mathematical analysis of human relationships.
Fig 1
In the sociogram, measuring the network location is finding the
centrality of a node. These measures help us to find the various roles and
grouping in a network, such as who are the connectors, leaders, bridges, isolates,
where are the clusters and who is in them, who is in the core of the network
and who is on the periphery?
Fig 2
Look at the social network in Fig 2. Two nodes are connected if they
interact in some way. We will use this network to explain three most popular
individual centrality measures: Degree Centrality, Closeness Centrality and Betweenness
Centrality.
Degree Centrality:
Degree centrality is the sum of all other actors who are directly
connected to the actor in concern. In Fig 2, D has the most direct connections
in the network, indicating that D is the most active node in the network. May be
we would like say the more connections, the better, especially in
personal network. But it is not always so. What really matters is where
those connections leads to not how many the connections are. In Fig 2, D
only connects to others in her clique where those are already connected to each
other.
Closeness Centrality:
Closeness centrality represents the mean of the geodesic distances
between some particular node and all other nodes connected with in. It can help
to understand as how long does it take for a message to spread inside the
network from a particular node. In Fig 2, although C and F have fewer
connections than D, the closeness centrality of them are higher than D, that is
say the pattern of their ties allow them to access all the nodes in the network
more quickly than D. Therefore, they are in a great position to get the best
visibility into what is happening in the network.
Betweenness Centrality:
Betweenness centrality counts the number of shortest paths between I and
k that actor j resides on. In the Fig 2, H has few direct connections, even
fewer than the average in the network. But, H is at one of the best locations
in the network——between
two important areaes. H plays a ‘broker’ or ‘bridge’ role in the network. A node
with high betweenness is able to act as a gatekeeper controlling the flow of
resources.
It is just a brief introduction of social network analysis. That’s all~
As you said in the blog,social network analysis focuses on the social actor and the relationship between the actors. So if we want to identify the position we state, we just try our best to analysis these relationship between the actors. To some extent, It's a easier way to gather different parameters such as degree centrality ,closeness centrality,between centrality and so on. When we see the sociogram,every user is a node, they connect with each other,how amazing it is!
回复删除yeah~ it is so amazing that sociogram convert the complex relationships into intuitive graph, which helps us to analyze those relationship with mathematical calculation.
删除It is a comprehensive explanation of the last class we have taken. To me, the sociogram is full of fun. I would really like to calculate my degree centrality ,closeness centrality or between centrality. Those classified data may help me collect information from sns much more efficient.
回复删除Actually when the situation comes to a directional Social Network Service, SNA will be more complicated even some of the algorithm will not change, because we shoule consider both the in-degree and out-degree of a node (the two degrees have different meanings). In addition, we can still find some ways to improve the result of SNA, i.e. Set different values for a relation in the SNS, range from 0 -1 , 0 represent no relation and 1 represent strongest relations.
回复删除yeah, good point. As you said, setting different values for a relation is helpful to improve the result of SNA. You know the degree of relation is not a simple linear relationship, it may use log or other mathematic method to calculate more appropriate degree of relationship.
删除You've really made a good explanation about these three most popular individual centrality measures,which helps us understand them better.Social network analysis is necessary for us nowadays,as wrote in your blog,it doesn't matter what you know, but who you know.
回复删除While doing SNA, only one or two sides can't reflect the real connection between two nodes. You are giving me a more clear knowledge about different individual centrality measures.
回复删除It seems that you article focus on centrality of social network. It's a important attribute to study a social network or a node in it. In some case, the person with higher centrality is the social queen, which means she is a node that can influence most other nodes.
回复删除Very good explanation about the three types of centrality, it helps me understand them better. It will be good to introduce the concept of influence range while talking about closeness centrality.
回复删除The essay made a useful illustration with an example to explain 3 centrality measures in social network analysis. This makes people to understand more about them with figures, especially figure 2.
回复删除nice picture, this article really gives me a great explanation about the centrality, this concept made me confused before. you introduce the three things:degree centrality, closeness centrality and betweenness centrality, which help to calculate the relationship between the people(represented by nodes) to make the whole map clear. interesting!
回复删除此评论已被作者删除。
回复删除Thanks for your comprehensive explanation of these three concepts, degree centrality, closeness centrality and between centrality. Actually, these concepts really make me confused for a time. Your blog also make me realize that the actor in social network can also be a group, an organization and even a country. What I always believe the actor is just people. Thank you and looking forward to your further work.
回复删除Very detail explanation of the concepts!
回复删除A good article helping me review these important concepts. The illuminations with figures point out their difference clearly.
回复删除