A New Measure Summarising ‘Information’ Conveyed in Cluster Analysis of Card-Sort Data:
Download
1 / 15

ECAI 2002: IDAMAP - PowerPoint PPT Presentation


  • 101 Views
  • Uploaded on

A New Measure Summarising ‘Information’ Conveyed in Cluster Analysis of Card-Sort Data: Application to a Neonatal Intensive Care environment G. Ewing , R. Logie, J. Hunter, N. McIntosh, S. Rudkin, Y. Freer and L. Ferguson. ECAI 2002: IDAMAP. Context.

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about ' ECAI 2002: IDAMAP' - cynthia-woodward


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

A New Measure Summarising ‘Information’ Conveyed in Cluster Analysis of Card-Sort Data:Application to a Neonatal Intensive Care environmentG. Ewing, R. Logie, J. Hunter, N. McIntosh, S. Rudkin, Y. Freer and L. Ferguson

ECAI 2002: IDAMAP


Context
Context Cluster Analysis of Card-Sort Data:

  • Initial effort in project, NEONATE, to develop decision support for clinical staff;

  • How can computer-based decision support improve patient care in neonatal intensive care units.


Elicitation of terms
Elicitation of Terms Cluster Analysis of Card-Sort Data:

  • Need for an agreed lexicon of terms to describe the actions carried out and the relevant sensory data (descriptors).

  • Desirable to investigate relationships between terms for optimal data input.

  • Interviews to elicit clinical actions (~70) and patient descriptors (552).

  • Subjects (32) : 8 JN, 8 SN, 8JD, 8 SD.

  • SME vetted actions (51) and descriptors lists (132);

    • Removal of singletons and synonyms.


Actions
Actions Cluster Analysis of Card-Sort Data:


Card sorting
Card-Sorting Cluster Analysis of Card-Sort Data:

  • "Concept Sorting" is a well-known technique.

  • A flexible method to label the basic categories and to reach an agreement on a conceptual structure.

  • SME sorts cards into piles (values) according to different criteria (attributes).

  • Cognitive psychology studies have shown that this is very efficient elicitation technique.


Card sorting procedure
Card-Sorting Procedure Cluster Analysis of Card-Sort Data:

Merge label

cards into

higher level

categories

Continue process until

highest level categories

reached.

Sort cards into piles

and label piles

Shuffled Set of Cards


Card sorting experiments
Card-Sorting Experiments Cluster Analysis of Card-Sort Data:

  • “Actions” card-sorts: Subjects (32) : 8 JN, 8 SN, 8JD, 8 SD (completed),

    • 2 sorts per subject (at different times);

    • Approx. I hours per session.

  • “Descriptor” card-sorts: Subjects (32) : 8 JN, 8 SN, 8JD, 8 SD (in progress, 1st sort 2/3 complete),

    • 2 sorts per subject (at different times)’

    • Approx. I.5 hours per session;


Analysis of results
Analysis of Results Cluster Analysis of Card-Sort Data:

  • Card-sorts were analysed by cluster analysis;

  • Dendrograms produced;

  • Summary graphs were developed;

  • Information measures were developed.


Dendrogram jn
Dendrogram JN Cluster Analysis of Card-Sort Data:

“Actions” dendrogram for junior nurses.


Dendrogram sd
Dendrogram SD Cluster Analysis of Card-Sort Data:

“Actions” dendrogram for senior doctors.


Graphical summary
Graphical Summary Cluster Analysis of Card-Sort Data:

Stylised Graphical Summary of Dendrogram Structure


Graphical summary 2
Graphical Summary 2 Cluster Analysis of Card-Sort Data:

Graphical Summary of (actual) Dendrogram Structure


Information channel
Information Channel Cluster Analysis of Card-Sort Data:

Channel matrix Q = qki

Alphabet, A = {ai}, N symbols

probability of ai = P(ai)

Alphabet, B = {bk}, M symbols

probability of bk = P(bk)


Results mutual information measures
Results: Cluster Analysis of Card-Sort Data:Mutual Information Measures


Questions
Questions? Cluster Analysis of Card-Sort Data:


ad