HarambeeNet Wiki

This wiki is for sharing ideas and resources about the science of networks in computer science education.

• HarambeeNet Project page
• Advisory Board Meeting Wiki
• Faculty Learning Community 2009-2010 Wiki
• Faculty Learning Community Wiki
• Resources to read for next Advisory Board Meeting
• 2009 Workshop page

Science of Networks courses Wiki details and categorizes the some example courses on the Science of Networks.

Science of Networks Courses Materials Wiki breaks down the textbooks and articles used in several Networks Classes.

- Science of Networks Course Modules Wiki gives a proposed module for a Science of networks Introductory Course at Duke University.

- Blog Project for Networks course

- Learning with Facebook Ideas for a Social Networking Module

News Articles on Networks in Education and Research Wiki

Duke Faculty and Publications Wiki

• Networks group at Notre Dame
• International Network for Social Network Analysis
• Collective Dynamics Group at Columbia led by Duncan Watts
• Supernetworks at UMass
• Weak Links group at Semmelweis University in Budapest, Hungary
• Reality Mining Machine Perception, Prediction and Learning of Complex Social Systems at MIT‘s Media Lab. Graphing of location via cellphone GPS and then predict future social interactions from that data (eg. If I ask Suzy for a date tonight, what is the probability she will accept?).
• Projects
  • Small Worlds Project at Columbia University
  • The Erdos Number Project
  • Oracle of Bacon
  • Cooperative Association for Internet Data Analysis: lots of data on physical and business connections of Internet
  • Modeling the Web: Steve Lawrence among others describe the link structure of the web and show that the existence of preferential attachment depends on the type of webpage. For example, creating a new popular web publication would be very difficult, while creating a new popular photography site would be much less so because of much more balanced growth.

• The Internet
• US Power Grid
• Interstate Highways

Among the most commonly modeled networks are acquaintanceship networks. In such a network, vertices are people in the world and edges denote some sort of acquaintance. These networks are often quite hard to measure because it is rather difficult to determine all of

One famous approximation was the Stanley Milgram‘s small world experiment. Letters were sent to 150 random people in Omaha, NE and Wichita, KS. The recipients were given the name and address of a target person in Cambridge, MA and instructed to send a letter to a friend who would be most likely to know the target person. Participants wrote their name on the letter to prevent loops and track the letter’s path. After a relatively small number of letters (5 median, 2-10 range), the letters arrived at their destination, leading some to hypothesize that all people in the world are connected by no more than 6 degrees of separation.

The Small Worlds Project tries to more precisely evaluate this 6 degrees hypothesis.

More examples of social networks include:

• Scientific collaboration (e.g Erdos numbers)
  • Vertices: math and computer science researchers
  • Edges: between coauthors on a published paper
• Movie collaboration (e.g. Oracle of Bacon)
  • Vertices: Actors and actresses
  • Edges: Appeared in a movie together

Jennifer Golbeck posted criteria for being a web-based social network. Popular examples include:

• The Facebook
• MySpace
• Friendster
• Black Planet
• Del.icio.us Social Bookmarking
• Flickr Social photo sharing and tagging
• Stickam Social multimedia sharing and tagging
• Frappr Social groups mapping and location
• Digg Social news sharing and ranking
• Sissy Fight The most ultimate and humiliating schoolyard popularity battle. Is it a social network? Is it a game?
• Friendster Pachinko Is it a social network? Is it a game? Are you willing to gamble with your friends?
• Running and Walking Routes Map and share running and walking routes. Great example of AJAX and useful user-created content.
• Electric Sheep Socially-designed digital art (the “flock” evolves to please its global audience).

• eBay bidding
• Corporate boards
  • Vertices: corporations
  • Edges: between companies that share a board member They Rule (Graph tool for visualizing cross pollination between corporate boards)
• Corporate partnerships
  • Vertices: corporations
  • Edges: represent formal joint ventures
• goods exchange networks
  • Vertices: buyers and sellers of commodities
  • Edges: represent “permissible” transactions

• Document similarity (see TouchGraph GoogleBrowser)
  • Vertices: documents on web
  • Edges: Weights defined by similarity
• Conceptual Network: Thesaurus like Wordnet
  • Vertices: words
  • Edges: synonym relationships

• Ecological networks (e.g. food chains)
• Disease spread networks
• Cellular networks
  • Vertices: substrate in organism
  • Edges: interactions between substrates

• Campaign donations
  • Huffington Post's Fundrace 2008
  • opensecrets.org
  • CQ MoneyLine
  • MAPLight.org
• Connections among politicians, lobbyists, fundraisers, etc.

• WordNet (see also WordNet related links
• Thinkmap Visual Thesaurus

• Committee on Network Science for Future Army Applications, National Research Council, Network Science, National Academies Press, 2007.

GUESS: extensible graph analysis package.

From the website:
GUESS is an exploratory data analysis and visualization tool for graphs and networks. The system contains a domain-specific embedded language called Gython (an extension of Python, or more specifically Jython) which supports the operators and syntactic sugar necessary for working on graph structures in an intuitive manner. An interactive interpreter binds the text that you type in the interpreter to the objects being visualized for more useful integration. GUESS also offers a visualization front end that supports the export of static images and dynamic movies.

• Manual (PDF)
• Adar, Eytan, GUESS: A Language and Interface for Graph Exploration (PDF), CHI 2006
• Python (Gython)
  • Instant Python: a primer on Python that gives some essentials
  • Python Tutorial: from Guido van Rossum, author of Python
  • Jython: a Python interpreter written in Java so that its easy to call Java code from Python and vice versa
  • Python Patterns - Implementing Graphs: a simple implementation of graphs in Python

• Taste: flexible collaborative filtering package for Java
  • Javadoc
• Duke Lifester: basic social network service based off of Lifester from the University of Pennsylvania’s Networked Life course
• Duke Scrobbler: music profile sharing tool based off of the AudioScrobbler service and the audioPod iPod plugin

• NING: Tools for building Web-based public or private social networks

• Definitions from Math World
• Graph Theory Tutorial: interactive tutorial with quizzes on graph terminology
• Graph Theory 3rd Edition (2005) by Reinhard Diestel: free online textbook

• GraphML
  • Quick GraphML guide
  • GraphML Primer
• DOT graph languages
  • Grappa 1.2: applet to draw DOT graphs
  • GraphViz: graph vizualizer
• GUESS .gdf format

• Visual Complexity Tools for visualizing networks, organized by network type (includes 44 social network graphing tools).
• Swivel: Social network for Viewing and discussing data visualizations, for viewing and discussing data sets, and for creating visualizations from existing data sets, including merging your data with others’ data.
• Many Eyes: IBM tool similar to Swivel.

• UCINET 6

Collaborative filtering algorithms make predictions (filtering) about the interest of a user on a new item based on the taste information from that user and other collaborating users from whom we have also collected taste information. The taste information consist of user preferences that either explicitly stated through some sort of ratings or inferred from the implicit preferences that are observed from normal user activity (purchases, music listened to, etc.).

• Collaborative Filtering Tutorial (PPT) by Prof. William Cohen of CMU Dept. of Machine Learning
• List of Collaborative Filtering Research Papers by James Thornton
• Recommender systems resources put together by Prof. Lyle Ungar of the University of Pennsylvania Dept. of Computer and Information Science
• The Baby Name Wizard A visualization tool for displaying the popularity of baby names by year.
• Good review papers
  • Herlocker, J, Konstan, J., Terveen, L., and Riedl, J. Evaluating Collaborative Filtering Recommender Systems. ACM Transactions on Information Systems 22 (2004), ACM Press, 5-53.
  • B. Sarwar, G. Karypis, J. Konstan and J. Riedl, Item-based collaborative filtering recommendation algorithms, in Proceedings of the Tenth International Conference on the World Wide Web (WWW 10), pp. 285-295, 2001.
  • J. Breese, D. Heckerman, C. Kadie. Empirical Analysis of Predictive Algorithms for Collaborative Filtering. In Proceedings of Fourteenth Conference on Uncertainty in Artificial Intelligence, Madison, WI, Morgan Kaufmann, July 1998. Also appears as Technical Report MSR-TR-98-12, Microsoft Research, May, 1998 (revised October, 1998).

• GroupLens Research Project at the University of Minnesota Dept. of Computer Science and Engineering
• Information Filtering Projects at MIT Media Lab
• Jon Herlocker’s Intelligent Information Systems group at Oregon State University Dept. of Computer Science

• LibraryThing‘s book recommendations and anti-recommendations
• Pandora: radio from the Music Genome Project

• MovieLens data set
• Book-Crossing data set
• Jester Joke data set
• Netflix Prize (movies) ====