Statistical Network Analysis:
Models, Issues, and New Directions

A Workshop at the
23rd International Conference on Machine Learning
(ICML 2006)

Thursday, June 29, 2006, Pittsburgh PA, USA




All sessions will be held in RANGOS #3, at the 2nd floor of the University Center (UC) building.


Program highlights

We welcome the following invited speakers.

  • David Krackhardt, Carnegie Mellon University   (web page)
  • Andrew McCallum, University of Massachusetts, Amherst   (web page)
  • Stanley Wasserman, Indiana University   (web page)

  Our program also features an invited panel discussion that will conclude the workshop (17:00-17:45). The panel brings together researchers from different communities, in order to foster a rich intellectual exchange, identify novel modeling approaches, diverse applications, and new research directions.




Proceedings

The workshop proceedgins will appear as a bound volume in the LNCS series published by Springer.




Schedule (ALL SESSIONS IN RANGOS #3, UC 2nd floor)

  8:00- 9:00 -- Breakfast

  9:00-10:30 -- Morning session I

    Invited talk: Heider vs Simmel: Comparing Generative Models of Network Formation (Abstract)
    . David Krackhardt (Carnegie Mellon University)

    A latent Space Model for rank data (PDF)
    . Isobel C. Gomley & Thomas B. Murphy (Trinity College Dublin)

    Exploratory study of a new model for evolving networks (PDF)
    . Anna Goldenberg & Alice Zheng (Carnegie Mellon University)

  10:30-11:00 -- Coffee

  11:00-12:30 -- Morning session II

    Invited talk: Latent variable models of social networks and text (Abstract)
    . Andrew McCallum (University of Massachusetts, Amherst)

    Approximate kalman filters for embedding author-word co-occurrence data over time (PDF)
    . Purnamrita Sarkar, Sajid M. Siddiqi & Geoffrey J. Gordon (Carnegie Mellon University)

    Analysis of a dynamic social network built from PGP keyrings (PDF)
    . Robert Warren, Dana Wilkinson (University of Waterloo) & Mike Warnecke (PSW Applied Research Inc.)

  12:30-14:00 -- Lunch & Poster Session -- joint with SRL & SOS workshops

    Stochastic block models of mixed membership: General formulation and "nested" variational inference (PDF)
    . Edoardo M. Airoldi (Carnegie Mellon University), David M. Blei (Princeton University), Stephen E. Fienberg & Eric P. Xing (Carnegie Mellon University)

    Exploratory study of a new model for evolving networks (PDF)
    . Anna Goldenberg & Alice Zheng (Carnegie Mellon University)

    A latent Space Model for rank data (PDF)
    . Isobel C. Gomley & Thomas B. Murphy (Trinity College Dublin)

    Information marginalization on subgraphs (PDF)
    . Jiayuan Huang, (University of Waterloo), Tingshao Zhu, Russel Greiner, Dale Schuurmans (University of Alberta) & Dengyong Zhou (NEC Laboratories America)

    Predicting protein-protein interactions using relational features (PDF)
    . Louis Licamele & Lise Getoor (University of Maryland, College Park)

    Age and geographic inferences of the LiveJournal social network (PDF)
    . Ian MacKinnon & Robert Warren (University of Waterloo)

    A brief survey of machine learning methods for classification in networked data and an application to suspicion scoring (PDF)
    . Sofus A. Macskassy (Fetch Technologies Inc.) & Foster Provost (New York University)

    Inferring formal titles in organizational email archives (PDF)
    . Galileo M.S. Namata Jr, Lise Getoor (University of Maryland, College Park) & Christopher P. Diehl (John Hopkins Applied Physics Laboratory)

    Approximate kalman filters for embedding author-word co-occurrence data over time (PDF)
    . Purnamrita Sarkar, Sajid M. Siddiqi & Geoffrey J. Gordon (Carnegie Mellon University)

    Discovering functional communities in dynamical networks (PDF)
    . Cosma R. Shalizi (Carnegie Mellon University) & Marcelo F. Camperi (University of San Francisco, San Francisco)

    Learning approximate MRFs from large transaction data (PDF)
    . Chao Wang & Srinivasan Parthasarathy (Ohio State University)

    Entity relationship labeling in affiliation networks (PDF)
    . Bin Zhao, Prithviraj Sen & Lise Getoor (University of Maryland, College Park)

  14:00-15:30 -- Afternoon session I

    Invited talk: A review of statistical models for networks (Abstract)
    . Stanley Wasserman (Indiana University)

    Discrete temporal models of social networks (PDF)
    . Steve Hanneke & Eric Xing (Carnegie Mellon University)

    A simple model for complex networks with arbitrary degree distribution and clustering (PDF)
    . Mark S. Handcock & Martina Morris (University of Washington, Seattle)

  15:30-16:00 -- Coffee

  16:00-16:30 -- Afternoon session II

    Strutural inference of hierarchies in networks (PDF)
    . Aaron Clauset, Cristopher Moore (University of New Mexico, Albuquerque) & Mark Newman (University of Michigan, Ann Arbor)

  16:30-18:00 -- Closing session

    Invited panel discussion
    . Chair: Stephen Fienberg (Carnegie Mellon University)   Panelists: David Blei (Princeton University), David Krackhardt (Carnegie Mellon University), Andrew McCallum (University of Massachusetts, Amherst), Cosma Shalizi (Carnegie Mellon University), Stanley Wasserman (Indiana University)

    Closing remarks




Overview

Many modern data analysis problems involve large data sets of artificial, social, and biological networks. In these settings, traditional IID assumptions are inappropriate; the analyses must take into account the structure of relationships between the data. As a result, there has been increasing research developing techniques for incorporating network structures into machine learning and statistics.

  Network modeling is an active area of research in several domains. Statisticians have mostly concentrated on models of static networks. These models are concerned with the existence of edges between individual nodes, but do not attempt to model aggregate properties. In contrast, physicists have addressed global properties of large complex networks. Their models describe average statistics of the network, or properties of typical networks in large ensembles; the links between particular nodes are less meaningful.

  This workshop focuses on probabilistic methods for network analysis, paying special attention to model design and computational issues of learning and inference. We are bringing together statistical network modeling researchers from different communities, thereby fostering collaborations and intellectual exchange. Our hope is that this will result in novel modeling approaches, diverse applications, and new research directions.




Organizers

  Edo Airoldi, Carnegie Mellon University
  David Blei, Princeton University
  Stephen Fienberg, Carnegie Mellon University
  Anna Goldenberg, Carnegie Mellon University
  Eric Xing, Carnegie Mellon University
  Alice Zheng, Carnegie Mellon University

Program Committee

  David Banks, Duke University
  Peter Dodds, Columbia University
  Lise Getoor, University of Maryland
  Mark Handcock, University of Washington, Seattle
  Peter Hoff, University of Washington, Seattle
  David Jensen, University of Massachusetts, Amherst
  Alan Karr, National Institute of Statistical Sciences
  Jon Kleinberg, Cornell University
  Andrew McCallum, University of Massachusetts, Amherst
  Foster Provost, New York University
  Cosma Shalizi, Carnegie Mellon University
  Padhraic Smyth, University of California, Irvine
  Josh Tenenbaum, Massachusetts Institute of Technology
  Stanley Wasserman, Indiana University