Lada Adamic, School of Information

Courses

SI 601 (F'07)
DRAT: data retrieval and analysis techniques
SI 508
(F'07)
Networks: Theory and Application (MS level)
SI 708 (W'O7,F'07)
Networks: Theory and Application (PhD level)
SI 503 (W'07)
Search and Retrieval
SI 544 (F'06)
Introduction to Statistics and Data Analysis
SI 614 (W'06) -> 508/708
Networks: Theory and Application

RIW Networks seminar (2006-2007)

Students
current

Xiaolin Shi (PhD)
Matt Bonner (CSE undergrad)

past
Noor Ali-Hasan (MSI - thesis '06)
Dale Hunscher (MSI - thesis '07)
Info on Lada

bio
CV
Code

demos in guess
scripts Matlab
demos NetLogo
image gallery
list of network software



NetLogo demos of network phenomena

 

These demos, created with the NetLogo software for building agent based models, cover the topics of network growth and diffusion. You can play with another set of demos, created using GUESS, that cover the topics of resilience/robustness, community structure, PageRank and search. Please feel free to use and modify any of the materials here.


Preferential vs. random attachment This model allows you to grow a network. Nodes arrive and attach to those already in the network. You control the parameter that determines whether they will prefer to attach to the "hubs" - the nodes that already have many connections, or if they will attach at random. Warning - you may observe some power law degree distributions.



Diffusion on a small world topology. This model allows you to construct small-world networks using the Watts-Strogatz model, and examine the effect the random rewirings (knowing people who are far away, and not just one's neighbors) has on disease propagation . Red nodes represent infected individuals, and you can tune the probability that the disease is transmitted across a tie from an infected individual to a non-infected contact during any given time period. This is a very slightly modified version of the SocialTies model.



More diffusion models:

Diffusion in an Erdos-Renyi graph (you get to control how big, how dense, and how infectious)

Diffusion in a BA graph (you get to control how big, how dense, how infectious, and how preferentially attaching)

 

Also check out the basic network models from the NetLogo folks themselves at Northwestern.