Visual Tool for Literature Exploration

1
Visual Tool for Literature Exploration

• Tingting Jiang
• November 14, 2006

2
Outline

• Literature Exploration
• Visualization Overview
• Visualization Applications
• Term Project

3
Literature Exploration

• Traditional activities in literature exploration
• Collecting
• Categorizing
• Reading
• Evaluating
• Writing

4
Literature Exploration Cont

• Collecting Literature Search
• 1. Identifying resources
• Databases, PittCat, E-journals
• Internet
• 2. Developing search strategies
• Keywords or phrases
• Broaden, narrow, or modify

5
Literature Exploration Cont

• Product of literature exploration literature
  review
• A literature review is a summary of
  previous research on a topic.

6
Literature Exploration Cont

• Questions to be answered in a literature review
• 1. What is known about the subject?
• 2. Are there any gaps in the knowledge of
  the subject?
• 3. Have areas of further study been
  identified by other researchers that you may want
  to consider?
• 4. Who are the significant research
  personalities in this area?
• 5. Is there consensus about the topic?
• 6. What aspects have generated significant
  debate on the topic?

7
Literature Exploration Cont

• Questions to be answered in a literature review
• 7. What methods or problems were identified
  by others studying in the field and how might
  they impact your research?
• 8. What is the most productive methodology
  for your research based on the literature you
  have reviewed?
• 9. What is the current status of research in
  this area?
• 10. What sources of information or data were
  identified that might be useful to you?

8
Visualization Overview

• Information visualization
• - The use of computer-supported, interactive,
  visual representations of abstract data to
  amplify cognition
• Knowledge visualization
• - The use of visual representations to
  transfer knowledge between at least two persons

9
Knowledge Visualization

• Purposes
• Reduce visual search time
• Comprehend large amounts of data
• Better understand complex data
• Identify key ideas, researchers, changes
  in a filed Knowledge transfer/Scholarly
  communication

10
Visual Representations

• Graphs (quantitative)
• Tables (words, numbers)
• Maps (spatial)
• Time charts (temporal)
• Network charts (node link)
• Diagrams (structure process)
• Icons
• Photos

11
Visualization Techniques

• Rearrangement
• A graphic is no longer drawn once for
  all it is constructed and reconstructed
  (manipulated) until all the relationships which
  lie within it have been perceived

12
Rearrangement Examples

• Table Lens

13
Visualization Techniques Cont

• Presentation
• Focus Context (Fisheye) researchers
  concentration on a problem can probably be
  enhanced if irrelevant detail are removed

14
Presentation Examples

• Perspective Wall
• Hyperbolic Tree (http//nsdl.org/browse/index.php)

15
Visualization Techniques Cont

• Interaction
• Overview
• Zoom
• Filter
• Details-on-demand
• Relate
• History
• Extract

16
Visualization Applications

• Dogpile (http//www.dogpile.com/)
• Vivisimo (http//vivisimo.com/)
• Clusty (http//clusty.com/)
• Grokker (http//www.grokker.com/)
• Mooter (http//www.mooter.com/)
• KartOO (http//www.kartoo.com/)
• ujiko (http//www.ujiko.com/)
• KwMap (http//www.kwmap.net/)
• TouchGraph (http//www.touchgraph.com/)
• RefViz (http//www.refviz.com/)

17
Dogpile

• InfoSpace, Inc.
• Metasearch engine Google, Yahoo! Search, MSN,
  Ask.com, About, MIVA, LookSmart and more
• Relevancy
• Metasearch technology ensuring best results top
  the list
• Missing Pieces visualization (disappear?)

18
Dogpile
19
Vivisimo

• Carnegie Mellon University 2000
• Award-winning search technology clustering
• Pre-retrieval Tagging vs. post-retrieval
  Clustering

20
Clusty

• Vivisimo 2004 Pittsburgh
• Metasearch engine Ask.com, MSN, Wikipedia, etc.
• Clusters
• Discover unexpected relationships between items
• Tree expand, contract

21
Grokker

• Groxis Inc.
• Metasearch engine Yahoo!, Wikipedia, Amazon
  Books
• Clusters
• Results grouped in topics rather than presented
  in a linear list where some results might be
  missed
• Outline View (tree) as well as Map View
  (interactive)

22
Grokker
23
Mooter

• Mooter Media 2003
• Clusters
• Node-link diagrams all the clusters separated on
  multiple pages

24
KartOO

• KartOO S.A.
• Metasearch engine
• Related topics help refine search
• Interactive cartographic maps one search
  generates several maps
• Lots of cool visual tricks not as relevant as
  expected

25
ujiko

• KartOO S.A.
• Sets of themes help improve search
• Interactive visualization
• Customizable search engine users decide the
  relevance of results
• The more you use it, the more functions it is
  able to offer

26
KwMap

• KwMap.Net
• Keyword search engine
• Refine search keywords related keywords and
  keywords variantions
• Two axes
• Results - websites

27
TouchGraph

• TouchGraph LLC
• Visualizations of associative networks
• Amazon browser, Google browser, and LiveJournal
  browser
• Interactive node-link diagrams
• Clusters

28
RefViz

• OmniViz Inc.
• A text analysis and visualization software
  application designed to retrieve, analyze,
  organize, and facilitate the comprehension of the
  huge amounts of literature
• Galaxy Matrix visualizations

29
RefViz - Galaxy

• Groups and references

30
RefViz - Matrix

• Groups and concepts

31
Summary

• Relevance vs. clustering
• Clusters classification vs. categorization
• Results content vs. sources
• Types textual vs. multimedia
• Keywords automatically vs. manually
• Browsing vs. searching
• Visualization features

32
Term Project

• Goal developing a new scheme for literature
  visualization (prototype or Web based system)
• Follow-up research developing a Web based tool
  for the whole process of literature exploration,
  not just collecting

33
Term Project

• Pre-visualization processing
• Classification schemes subjected to change
• Literature resources
• Collaborative human reading
• Filtering, tagging, and submitting to
  semi-hierarchy

34
Term Project

• Visualization highlights
• Complementary visualization views
  semi-hierarchical, time, and region
• Tag-oriented
• Tag to knowledge fraction mapping
• Browsing as well as searching
• Rearrangement, Presentation, and Interaction