InfoVis Cyberinfrastructure

1

• InfoVis Cyberinfrastructure
• Shashikant Penumarthy, Bruce Herr Katy Börner
• School of Library and Information Science
• sprao bherr katy _at_indiana.edu
• Workshop on Information Visualization Software
  Infrastructures
• IEEE InfoVis Conference, Austin, TX

2
InfoVis Cyberinfrastructure (IVC) Software
Framework

• II.1) Project Name and Web Address
• Information Visualization Cyberinfrastructure
  (IVC) Software Frameworkhttp//iv.slis.indiana.ed
  u/sw/http//sourceforge.net/projects/ivc/
• II.2) Core Team Members
• Developer, Shashikant Penumarthy, SLIS, Indiana
  University sprao_at_indiana.edu
• Developer, Bruce Herr, CS, Indiana University
  bherr_at_indiana.edu
• Project Manager, Katy Börner, SLIS, Indiana
  University katy_at_indiana.edu
• II.3) Project Start Date
• May 11, 2004. The IVC Software Framework
  supersedes the original Information Visualization
  Repository that we started in 2000.

3

• II.4) Targeted User Group
• Researchers interested in the generation of
  Knowledge Domain Visualizations (KDVis) -- see
  also PNAS 101 (Suppl. 1) Apr 6, 2004 on 'Mapping
  Knowledge Domains'.
• Researchers and educators to run and compare
  commonly used data cleaning, analysis, modeling,
  visualization and interactivity algorithms.
• Algorithm developers interested in using the IVC
  as a medium for making new algorithms freely
  available.
• II.5) Supported User Tasks
• Serialization of algorithms, e.g., network
  extraction (e.g., co-author networks from
  publication data) -gt network analysis (e.g.,
  identification of hubs) -gt network layout (e.g.,
  using Kamada Kawai, hubs are visually distinct)
  -gt network interaction (e.g., zoom).
• Algorithm comparison, e.g., display a directory
  hierarchy using a JTree, a radial tree, a
  hyperbolic tree or a treemap layout, to determine
  the usability of (new) algorithms or to
  teach/learn the (dis)advantages of different
  algorithms.
• Conversion between data types. (e.g., load a
  co-author matrix -gt convert to graph -gt
  prune/cluster the graph using an importance
  metric -gt visualize/save resultant graph to
  disk.)

4
(No Transcript)
5

• http//vw.indiana.
• edu/aag05/

6

• II.4) Targeted User Group
• Researchers interested in the generation of
  Knowledge Domain Visualizations (KDVis) -- see
  also PNAS 101 (Suppl. 1) Apr 6, 2004 on 'Mapping
  Knowledge Domains'.
• Researchers and educators to run and compare
  commonly used data cleaning, analysis, modeling,
  visualization and interactivity algorithms.
• Algorithm developers interested in using the IVC
  as a medium for making new algorithms freely
  available.
• II.5) Supported User Tasks
• Serialization of algorithms, e.g., network
  extraction (e.g., co-author networks from
  publication data) -gt network analysis (e.g.,
  identification of hubs) -gt network layout (e.g.,
  using Kamada Kawai, hubs are visually distinct)
  -gt network interaction (e.g., zoom).
• Algorithm comparison, e.g., display a directory
  hierarchy using a JTree, a radial tree, a
  hyperbolic tree or a treemap layout, to determine
  the usability of (new) algorithms or to
  teach/learn the (dis)advantages of different
  algorithms.
• Conversion between data types. (e.g., load a
  co-author matrix -gt convert to graph -gt
  prune/cluster the graph using an importance
  metric -gt visualize/save resultant graph to
  disk.)

7

• II.6) Major Features of the System Architecture
• The IVC is a pluggable framework. Each software
  component part of the IVC can be plugged-in or
• unplugged as needed. This way, new kinds of
  algorithms, but also new data structures, new
• persistence methods, new look and feels for the
  interface and even entire toolkits can be easily
• integrated.
• The IVC framework can be divided into the
  following components
• IVC Core This is the manager of all system
  components and resources. It comprises the
  registries, the initializer and the IVC class.
• Data Models All supported data structures fall
  into this category. Support for data models is
  determined completely by the installed plug-ins.
• Persistence This consists of pluggable
  components each of which provides ways to store
  data to disk or other locations such as a
  database and to load data into the data models.
• Graphical User Interface (GUI) The menu driven
  front-end for the system.
• Plug-Ins The analysis, modeling and
  visualization algorithms or toolkits.
• Scheduler - Enables a user to set up a sequence
  of analyses or set up conditional execution
  pipelines.
• Logger - Logs user actions so that one can go
  back and see the sequence of steps one took to
  get to a particular result starting from a
  particular dataset.

8

• II.6) Major Features of the System Architecture

9

• II.7) Algorithms Provided
• The following algorithms are currently available
• Preprocessing Stop Word Removal, Porter
  Stemmer, Term Document Matrix.Modeling  Random
  Networks, Watts Strogatz, Barabasi Albert models,
  P2P Networks, (CAN, CHORD, Pru,
  Hypergrid),TARLAnalysis Burst Detection,
  Betweenness Centrality Clustering, ABSURDIST
  (concept-matching)Search Random-Walk, BFS,
  Search in P2P NetworksNetwork Layouts
  Fruchterman-Rheingold, Kamada-Kawai, Spring,
  Circle.Interaction Distortion, FisheyeMenu,
  ZoomingPan (Prefuse demos)Toolkits AW Toolkit
  (Analysis and Visualization of Virtual World
  Behaviour), Network Analysis Toolkit
• The following algorithms are currently being
  integratedAnalysis Ward Clustering, Path
  Finder Network Scaling, Similarity Flooding,
  Latent Semantic Analysis, Vector Space Model,
  SimVis (Matrix visualization)Visualization
  BurstVis (bursty events over time), Radial Tree,
  Hyperbolic Tree, Treemap.

10

• Demo

11

• Part III Main interest in participating in the
  workshop
• Determine the feasibility of combining efforts to
  create a common, shared IV infrastructure as
  opposed to 100s of underfunded or proprietary
  toolkits, platforms and frameworks.
• Scouring for ideas for a common data protocol for
  communication between plugins.
• Eliciting feedback about the IVC software
  architecture with regard to extensibility and
  ensuring that it is future-proof.