SlideShare a Scribd company logo
Knowledge Elicitation
Concept Sorting
Knowledge Elicitation (Introduction)
• Knowledge Elicitation is the process of acquiring knowledge about a
specific domain.
• It is one of the most important and a crucial task of the development of an
expert system since it directly has an impact on the overall quality of the
system.
• Knowledge elicitation is also often viewed as the bottleneck in the
development of expert systems or knowledge based systems as it is difficult
and time consuming activity.
• Among various known Knowledge Elicitation techniques available choice to
be used depends on the nature of the situation within which the
knowledge is elicited, the domain knowledge and availability of experts.
Knowledge Elicitation (Introduction) Cont…
• Knowledge Elicitation process gets tricky as the vast amount of information
is often kept inside the heads of domain experts.
• This makes the entire process complicated as the domain experts may not
be willing to disclose the information, due to worries of being sidelined or
becoming less important or getting redundant.
• There are various techniques used in the Knowledge Elicitation process :
• Interview
• Protocol Analysis
• Laddering
• Concept Sorting (Will be discussing)
• Repertory Grid
• Structural Assessment
When to use which technique?
• For Knowledge identification
• Unstructured interview, laddering
• For Knowledge specification
• Domain schema: Concept Sorting, repertory grid
• Template selection: self report
• Task & inference knowledge: self report
• For Knowledge Refinement
• Structured interview
What is Concept Sorting ?
• Concept sorting is a psychological technique that is useful in tapping
organization knowledge.
• It a way of finding out how an expert compares, orders concepts and relates
among a set of concepts.
• Used to capture conceptual knowledge and tacit knowledge.
• Can also sort objects or pictures instead of cards
• It works best on small group.
• Simple to apply
• Its is also Called Card Sorting.
• Simplest form is Concept Sorting
• Collection of concepts (or other knowledge objects) are written on separate cards
• Cards sorted into piles by an expert in to piles - each card in a pile must have something
in common
• Each time the cards are sorted it will be based on an attribute and each pile will
represent a value
Concept Sorting – How To ?
• To apply this technique , the KE follows the following steps:
1- First , KE Decide what classes of concepts he/she want to explore
(in particular their properties – attributes and values).
2- Consults a textbook, training manual, or in-house domain expert to
identify the major top-level concepts represented in the domain.
3- Place write the name of each concept on a separate note card .
4- Next, the KE asks the domain expert to begin sorting these cards
placing them in groups according to those that belong together.
5- As the domain expert sorts the cards the KE uses questioning
techniques to determine why they are placed together.
6- Repeat steps 4, and 5 until the expert can’t sort anymore
Example 1: Classification based of Habitat
Star
Fish
Bear
Cat Shark
Deer
Seahorse
Crab
Crocodile
Frog
Elephant
Dolphin
Example 1: Classification based of Habitat
Cont…
Dolphin
Aquatic Animals
ns
Terrestrial Animal
Star
Fish
Bear
Cat
Shark
Elephant
Deer
Seahorse
Crab
Crocodile
Frog
Amphibia
Card Habitat Size Eating Habit Vertebrate / Invertebrate
Mammals/
Non Mammals
Elephant 1 3 1 1 1
Cat 1 1 3 1 1
Deer 1 2 1 1 1
Bear 1 3 3 1 1
Crab 2 1 2 2 2
Crocodile 2 3 2 1 2
Frog 2 1 2 1 2
Dolphin 3 2 3 2 1
Starfish 3 1 3 2 2
Sea Horse 3 1 3 1 2
Shark 3 3 2 1 2
1-Teresstrial 1-Small 1-Herbivorous 1-Vertebrate 1-Mammals
2- Amphibia 2-Medium 2-Carnivorous 2-Invertebrate 2-Non-mammal
3-Aquatic 3-Large 3-Omnivorous
Example 2
• The Number of 15 models of cars may be grouped into two categories named “Foreign"
and “Domestic
• Then re-sorted into three categories; “Sedan" , “Hatchback", and “Sports";
• Then re-sorted into four categories; " Expensive Domestic", " Less Expensive Domestic",
" Expensive Foreign", “Less Expensive Foreign“.
• A hierarchy tree may be created expressing the group categories as levels within the
hierarchy.
Based on Cost
Based on Design and
Performance
Based on Design and
Performance
Based on Cost
Example 2 Cont…
Domestic
Cars
Foreign
Less Expensive
Expensive SedanHatchback Sports
Less Expensive
ExpensiveSedanHatchback Sports
Application of Concept Sorting
• Concept sorting is basically used in several disciplines like
- Knowledge Engineering
- Psychology
- Marketing etc.
Pros and Cons
• Pros
• Fast to apply and easy to analyze
• Often instructive to the expert in a sense that it may lead expert to see
structure that he himself has not consciously articulated before
• Time saving by not having to transcribe and analyze lengthy verbal reports
• Can be used for images and objects as well
Pros and Cons Cont…
• Cons
• Experts can often confound dimensions by not consistently applying the same
semantic distinctions throughout an elicitation session
• May face problem in categorization of elements in meaningful way
Conclusion
• Card Sorting techniques provide a means of achieving a more focused
or systematic understanding of the classifications and relationships in
the expert's domain.
Reference
http://liris.cnrs.fr/amille/enseignements/MasterCode/IC_IA/Session1_
2008/commonkads/06-km-process.ppt
http://www.cse.aucegypt.edu/~rafea/csce485es/slides/ch.10.pdf
http://studentnet.cs.manchester.ac.uk/ugt/COMP34512/slides/day2.p
df
http://academic.cankaya.edu.tr/~agorur/AI/lect/knowledge.html

More Related Content

PPTX
Java Constructors
PPTX
Multi Processors And Multi Computers
PPTX
Threading in C#
PPTX
Dijkstra’s algorithm
PDF
Algorithms Lecture 4: Sorting Algorithms I
PPTX
Dijkstra's Algorithm
PDF
P, NP, NP-Complete, and NP-Hard
PPTX
Daa unit 1
Java Constructors
Multi Processors And Multi Computers
Threading in C#
Dijkstra’s algorithm
Algorithms Lecture 4: Sorting Algorithms I
Dijkstra's Algorithm
P, NP, NP-Complete, and NP-Hard
Daa unit 1

What's hot (20)

DOCX
Java questions for viva
PPTX
Syntax-Directed Translation into Three Address Code
PPT
12 multi-threading
 
PPT
CSharp.ppt
PPTX
Dijkstra's Algorithm
PPTX
unit 1 ppt.pptx
PDF
Java Course 11: Design Patterns
PPTX
Solid design principles
PPTX
Constructor in java
PPT
Priority scheduling algorithms
PPTX
Tic tac toe game with graphics presentation
PPT
Cocomo models
PPTX
Uni Processor Architecture
PDF
LeetCode Solutions In Java .pdf
PDF
Java Programming :Event Handling(Types of Events)
PPTX
Flutter
PDF
Compiler design lab programs
PPTX
Directed Acyclic Graph Representation of basic blocks
PPTX
Control Statements in Java
PPTX
Merge sort analysis and its real time applications
Java questions for viva
Syntax-Directed Translation into Three Address Code
12 multi-threading
 
CSharp.ppt
Dijkstra's Algorithm
unit 1 ppt.pptx
Java Course 11: Design Patterns
Solid design principles
Constructor in java
Priority scheduling algorithms
Tic tac toe game with graphics presentation
Cocomo models
Uni Processor Architecture
LeetCode Solutions In Java .pdf
Java Programming :Event Handling(Types of Events)
Flutter
Compiler design lab programs
Directed Acyclic Graph Representation of basic blocks
Control Statements in Java
Merge sort analysis and its real time applications
Ad

Similar to Knowledge Elicitation Techiniques Concept Sorting (20)

PPTX
Concept Sorting in Knowledge Elicitation
PPT
2004 06 intelligence analysis seminar
PPTX
Chapter 03km
PDF
Jeff Lopez - To Affinity and Beyond
PDF
Jeff Lopez - To Affinity and Beyond
DOCX
!#$&()&#+,$)!#$$&())• +,-.$0$12,#-34-$#3.docx
KEY
Game Design 2: Expert Evaluation of User Interfaces
PPTX
Information Architecture. Card Sorting
PPTX
Expert systems
PPT
KAI, the Information Specialist
PPTX
Introduction to Machine Learning
PPT
data analysis.ppt
PPTX
data analysis.pptx
PPTX
The Art of Project Estimation
PDF
Team dynamics: The Joys and Sorrows of Diverse Teams by Rebecca Parsons, CTO,...
PPTX
Using an Agile Inception to Kick Off a Project
PPTX
eSource Stakeholders Group 18mar2016
PDF
Managing your Metadata w/ SharePoint 2010
PPTX
Requirements Management Part 1 - Management and Elicitation
Concept Sorting in Knowledge Elicitation
2004 06 intelligence analysis seminar
Chapter 03km
Jeff Lopez - To Affinity and Beyond
Jeff Lopez - To Affinity and Beyond
!#$&()&#+,$)!#$$&())• +,-.$0$12,#-34-$#3.docx
Game Design 2: Expert Evaluation of User Interfaces
Information Architecture. Card Sorting
Expert systems
KAI, the Information Specialist
Introduction to Machine Learning
data analysis.ppt
data analysis.pptx
The Art of Project Estimation
Team dynamics: The Joys and Sorrows of Diverse Teams by Rebecca Parsons, CTO,...
Using an Agile Inception to Kick Off a Project
eSource Stakeholders Group 18mar2016
Managing your Metadata w/ SharePoint 2010
Requirements Management Part 1 - Management and Elicitation
Ad

Recently uploaded (20)

PDF
Soil Improvement Techniques Note - Rabbi
PDF
Unit I ESSENTIAL OF DIGITAL MARKETING.pdf
PPTX
UNIT 4 Total Quality Management .pptx
PDF
The CXO Playbook 2025 – Future-Ready Strategies for C-Suite Leaders Cerebrai...
PDF
Integrating Fractal Dimension and Time Series Analysis for Optimized Hyperspe...
PDF
BIO-INSPIRED ARCHITECTURE FOR PARSIMONIOUS CONVERSATIONAL INTELLIGENCE : THE ...
PDF
Visual Aids for Exploratory Data Analysis.pdf
PPTX
Nature of X-rays, X- Ray Equipment, Fluoroscopy
PDF
Artificial Superintelligence (ASI) Alliance Vision Paper.pdf
PDF
PPT on Performance Review to get promotions
PPTX
Artificial Intelligence
PDF
Mitigating Risks through Effective Management for Enhancing Organizational Pe...
PPT
introduction to datamining and warehousing
PPTX
MET 305 2019 SCHEME MODULE 2 COMPLETE.pptx
PDF
null (2) bgfbg bfgb bfgb fbfg bfbgf b.pdf
PDF
R24 SURVEYING LAB MANUAL for civil enggi
PDF
86236642-Electric-Loco-Shed.pdf jfkduklg
PDF
737-MAX_SRG.pdf student reference guides
PPT
Introduction, IoT Design Methodology, Case Study on IoT System for Weather Mo...
PDF
BIO-INSPIRED HORMONAL MODULATION AND ADAPTIVE ORCHESTRATION IN S-AI-GPT
Soil Improvement Techniques Note - Rabbi
Unit I ESSENTIAL OF DIGITAL MARKETING.pdf
UNIT 4 Total Quality Management .pptx
The CXO Playbook 2025 – Future-Ready Strategies for C-Suite Leaders Cerebrai...
Integrating Fractal Dimension and Time Series Analysis for Optimized Hyperspe...
BIO-INSPIRED ARCHITECTURE FOR PARSIMONIOUS CONVERSATIONAL INTELLIGENCE : THE ...
Visual Aids for Exploratory Data Analysis.pdf
Nature of X-rays, X- Ray Equipment, Fluoroscopy
Artificial Superintelligence (ASI) Alliance Vision Paper.pdf
PPT on Performance Review to get promotions
Artificial Intelligence
Mitigating Risks through Effective Management for Enhancing Organizational Pe...
introduction to datamining and warehousing
MET 305 2019 SCHEME MODULE 2 COMPLETE.pptx
null (2) bgfbg bfgb bfgb fbfg bfbgf b.pdf
R24 SURVEYING LAB MANUAL for civil enggi
86236642-Electric-Loco-Shed.pdf jfkduklg
737-MAX_SRG.pdf student reference guides
Introduction, IoT Design Methodology, Case Study on IoT System for Weather Mo...
BIO-INSPIRED HORMONAL MODULATION AND ADAPTIVE ORCHESTRATION IN S-AI-GPT

Knowledge Elicitation Techiniques Concept Sorting

  • 2. Knowledge Elicitation (Introduction) • Knowledge Elicitation is the process of acquiring knowledge about a specific domain. • It is one of the most important and a crucial task of the development of an expert system since it directly has an impact on the overall quality of the system. • Knowledge elicitation is also often viewed as the bottleneck in the development of expert systems or knowledge based systems as it is difficult and time consuming activity. • Among various known Knowledge Elicitation techniques available choice to be used depends on the nature of the situation within which the knowledge is elicited, the domain knowledge and availability of experts.
  • 3. Knowledge Elicitation (Introduction) Cont… • Knowledge Elicitation process gets tricky as the vast amount of information is often kept inside the heads of domain experts. • This makes the entire process complicated as the domain experts may not be willing to disclose the information, due to worries of being sidelined or becoming less important or getting redundant. • There are various techniques used in the Knowledge Elicitation process : • Interview • Protocol Analysis • Laddering • Concept Sorting (Will be discussing) • Repertory Grid • Structural Assessment
  • 4. When to use which technique? • For Knowledge identification • Unstructured interview, laddering • For Knowledge specification • Domain schema: Concept Sorting, repertory grid • Template selection: self report • Task & inference knowledge: self report • For Knowledge Refinement • Structured interview
  • 5. What is Concept Sorting ? • Concept sorting is a psychological technique that is useful in tapping organization knowledge. • It a way of finding out how an expert compares, orders concepts and relates among a set of concepts. • Used to capture conceptual knowledge and tacit knowledge. • Can also sort objects or pictures instead of cards • It works best on small group. • Simple to apply • Its is also Called Card Sorting. • Simplest form is Concept Sorting • Collection of concepts (or other knowledge objects) are written on separate cards • Cards sorted into piles by an expert in to piles - each card in a pile must have something in common • Each time the cards are sorted it will be based on an attribute and each pile will represent a value
  • 6. Concept Sorting – How To ? • To apply this technique , the KE follows the following steps: 1- First , KE Decide what classes of concepts he/she want to explore (in particular their properties – attributes and values). 2- Consults a textbook, training manual, or in-house domain expert to identify the major top-level concepts represented in the domain. 3- Place write the name of each concept on a separate note card . 4- Next, the KE asks the domain expert to begin sorting these cards placing them in groups according to those that belong together. 5- As the domain expert sorts the cards the KE uses questioning techniques to determine why they are placed together. 6- Repeat steps 4, and 5 until the expert can’t sort anymore
  • 7. Example 1: Classification based of Habitat Star Fish Bear Cat Shark Deer Seahorse Crab Crocodile Frog Elephant Dolphin
  • 8. Example 1: Classification based of Habitat Cont… Dolphin Aquatic Animals ns Terrestrial Animal Star Fish Bear Cat Shark Elephant Deer Seahorse Crab Crocodile Frog Amphibia
  • 9. Card Habitat Size Eating Habit Vertebrate / Invertebrate Mammals/ Non Mammals Elephant 1 3 1 1 1 Cat 1 1 3 1 1 Deer 1 2 1 1 1 Bear 1 3 3 1 1 Crab 2 1 2 2 2 Crocodile 2 3 2 1 2 Frog 2 1 2 1 2 Dolphin 3 2 3 2 1 Starfish 3 1 3 2 2 Sea Horse 3 1 3 1 2 Shark 3 3 2 1 2 1-Teresstrial 1-Small 1-Herbivorous 1-Vertebrate 1-Mammals 2- Amphibia 2-Medium 2-Carnivorous 2-Invertebrate 2-Non-mammal 3-Aquatic 3-Large 3-Omnivorous
  • 10. Example 2 • The Number of 15 models of cars may be grouped into two categories named “Foreign" and “Domestic • Then re-sorted into three categories; “Sedan" , “Hatchback", and “Sports"; • Then re-sorted into four categories; " Expensive Domestic", " Less Expensive Domestic", " Expensive Foreign", “Less Expensive Foreign“. • A hierarchy tree may be created expressing the group categories as levels within the hierarchy.
  • 11. Based on Cost Based on Design and Performance Based on Design and Performance Based on Cost Example 2 Cont… Domestic Cars Foreign Less Expensive Expensive SedanHatchback Sports Less Expensive ExpensiveSedanHatchback Sports
  • 12. Application of Concept Sorting • Concept sorting is basically used in several disciplines like - Knowledge Engineering - Psychology - Marketing etc.
  • 13. Pros and Cons • Pros • Fast to apply and easy to analyze • Often instructive to the expert in a sense that it may lead expert to see structure that he himself has not consciously articulated before • Time saving by not having to transcribe and analyze lengthy verbal reports • Can be used for images and objects as well
  • 14. Pros and Cons Cont… • Cons • Experts can often confound dimensions by not consistently applying the same semantic distinctions throughout an elicitation session • May face problem in categorization of elements in meaningful way
  • 15. Conclusion • Card Sorting techniques provide a means of achieving a more focused or systematic understanding of the classifications and relationships in the expert's domain.