CASOS 2007 Summer Institute

June 25 - July 1, 2007

  • Where?
    • Pittsburgh, PA, Carnegie Mellon University, Robert Engineering Hall - Singleton Room
  • When?
    • Monday June 25 - Sunday July 1, 2007
  • Who participates? Participation is open to graduate students, faculty, and personnel from industry and government. Due to space restrictions we are limited to approximately 35 attendants per year.
    Registration Fee:
    • CMU Students: $375
    • Grad Students: $675
    • Faculty: $950
    • Government and Industry: $1300
  • Please note, we accept checks, VISA, MasterCard and American Express for payment. Please make checks out to Carnegie Mellon University.
    For all payments contact Rochelle Economou.
  • Attention: If you need to cancel your registration, you must do so by close of business (5:00 pm, EDT) today, Friday, June 8, 2007 to receive a full refund. If you cancel after Friday, June 8, 2007, you will not receive a refund. We will not be able to refund your registration fee after Friday June 8, 2007.
  • Bring your own machine
  • You are strongly encouraged to bring your own laptop. However, you must have it uptodate in terms of security features, and it must be registered. Please see the MAC Address registration link below.

    Warning: Please make sure that you laptop is running Windows. The software will not run under VISTA. If bringing a MAC OS laptop please load Windows not all the software will run on a Mac OS.
  • Bring your own data
  • The CASOS group invites you to bring your own data. If you do not have your own data, don't worry, we will provide some. Here are examples of types of data that you might want to bring:

      1. Texts in .txt format that they want to run through AutoMap.
      Examples are: email messages, webpage content, paper abstracts, news articles, comment fields from fixed formatted files.

      Restrictions: Put one message, page, abstract, article per file.

      Put all files for same group or time period in a folder. Exclude pictures.

      2. Some type of relational data in DL, dynetml, csv, excel, or other standard network format. Data can also be in mysql or access if you know how to write queries to extract it.

    Examples are: who talks to whom, trade level between countries, semantic networks, event networks.

    Restrictions: Look at CASOS tools web page for easy formats to read. If you have node by attribute data, such as for each person degree, age, position you can use that data to create relational data.


    Make sure your machine is large enough for your data. You can have multiple networks. Each network can be in its own file, but need not be. If you are bringing your own machine, which we highly recommend, just bring your data on you machine.
  • MAC Address Registration

  • Everyone bringing their laptop (MAC or PC) must provide this information.
    Please email your info to Ed Walter.

  • Where to stay?
  • Agenda*Now Available!*
  • CASOS Summer Institute Outing

  • The Pittsburgh Pirates, College Night with the Pittsburgh Pirates
    Friday, June 29, 7:05 pm vs. Washington
    Rain night will be Saturday June 30.
    Ticket price is $20 per person. This includes a free Pirates t-shirt, food and drink vouchers.
    If you are interested, please contact Rochelle Economou by May 23 to confirm your spot. We will be taking payment for the tickets at a later date.
    We will be able to take last minute people on the first day of the institute, but we encourage you to let us know in advance!
    Terrill Frantz will tbe the Scene Coordinator for this event. Please contact him with any logistics questions.

  • Campus Map
  • Reading List *Now Available!*
  • (Please contact Rochelle for access.)
  • What to Bring
  • Commonly Asked Questions
  • Logistics
  • What to do in Pittsburgh
  • Purpose:

    The purpose of the CASOS summer institute is to provide an intense and hands-on introduction to dynamic network analysis and computational modeling of complex socio-technical systems. Both network analysis and multi-agent modeling will be covered. Participants will be able to complete the institute without programming skills or in-depth understanding of particular social theories. Computer programming and basic social or organizational theory are not included under the topics covered.

    Participants learn about current trends, practices, and tools available for social networks analysis, link analysis, simulation, and multi-agent modeling. Basic social network and dynamic network representations, statistics, analysis and visualization techniques are covered.Techniques for designing, analyzing, and validating computational models with and without network components are presented. There is also an emphasis on appropriate and inappropriate ways to critique computational models and network analyses. The strengths and weaknesses of computational and network approaches to examining complex socio-technical issues are discussed. Multiple computational platforms are explored and hands-on experience are provided.An examination of social network methods, complexity theory and procedures for integrating network-based metrics and statistics into computational models completes the program.The platforms students will work with include ORA, AutoMap, Construct, and UCINET.

    Students are encouraged to bring their own data and to learn to use these tools to code, analyze, reason about and visualize there data. Students will work through a tool chain where they extract networks from texts, analyze those networks, and the using simulation techniques evolve those networks.

    Institute Curriculum:

    The curriculum builds on both social network and computational analysis techniques, and illustrates how to use these techniques to study social, organizational and policy issues.

    Topics covered include:
    • Social Network Analysis
    • Dynamic Network Analysis
    • Elementary link analysis
    • Network based grouping techniques - Cliques, fuzzy groups, newman-girvan
    • Network based text analysis techniques
    • Basic Approach to building and evaluating multi-agent simulation systems
    • Basic machine learning/optimization techniques for use in multi-agent simulation - e.g., simulated annealing
    • Adaptive, Evolutionary, and Learning Systems
    • Validation and docking
    Faculty:
    • David Krackhardt
    • Kathleen M. Carley
    • Lee Wagenhalls