This document provides an overview of social network analysis, including what social networks are, what can be learned from analyzing social networks, and how social network analysis can be performed. Some key findings that can be uncovered include the six degrees of separation principle, the 80-20 rule of social popularity where a minority of nodes have most connections, how to identify influential nodes, and how to group similar nodes into communities. Various metrics and models are described for analyzing features like path lengths, degree distributions, ranking nodes, measuring community structure, and more. Examples of social network analysis are also provided.
74. Barabasi-Albert model
When a new node joins the network, it
connects to popular existing nodes
The probability is proportional to the existing
node’s number of links
119. Largest groups
Group 1
- 395 females, 201 males
- Average age: 42.4
Group 2
- 228 females, 83 males
- Average age: 38.9
Group 3
- 51 females, 27 males
- Average age: 34.1
120. Group 1 vs 2: p-value = 0.001
Group 1 vs 3: p-value < 0.001
Group 2 vs 3: p-value < 0.001
Thank you for the intro, I’m Han, very glad to be part of this great tradition, share a few exciting things that I know about sna
While you are working on the sandwich, let me share two examples that demonstrate the power of social network analysis
This network is about people, but the relationship is their board membership. If you think money is their friendship
Not only colorful but also very tasteful network
This network is about papers
What knowledge have we gained? Social network is a complex system where patterns only emerge on a large scale
Powerful tools to understand the rich and complex structures
We feel the world is small, your friend posts a message and another friend replies, your best friend from primary school is marrying your college roommate
Inspired by the fact that people move around and form new connections
In the first 17 years of my life…
We already have theories, what do we need next? Experiments!
We already have theories, what do we need next? Experiments!
Bacon started a charitable organization, sixdegrees.org
Bacon started a charitable organization, sixdegrees.org
Me – my wife – Prof. Mirkin – President Obama
This is where statistics come into play
Erdos-Renyi network
There is no such a number as scale
Prob to connect is not the same
NetLogo example
NetLogo example
NetLogo example
NetLogo example
A friend of mine, now a professor at IU
So I took the inspiration and made a perhaps more exciting example
No information on the airport itself
Actually shot in Evanston, my friend saw Matt Damon. The world is indeed very small!
We are not Facebook or LinkedIn, social network is not our product
If you are interested in the general idea of network science, read Linked. If you are eager to see the math behind it, read Newman’s book
Finally, I work at Quantum Lead, and we do data-driven sales optimization. We are located at the SW corner of 6th floor, near Fish Bowl. Come to talk to us if you are interested, we have a very talented team, awesome engineers and awesome managers.
With that, I will end my talk today. Thanks very much for your attention and please let me know if you have any questions!