Where might knowledge reside En coded knowledge Text programming Em brained knowledge CEO chess player fire fighter Em bodied knowledge Learning by doing learning occurs not just in the brain body memory ID: 628743
Download Presentation The PPT/PDF document "Knowledge Management Knowledge modeling ..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Knowledge Management
Knowledge modeling and concept map Slide2Slide3
Where might knowledge reside?
“En
coded”
knowledge
Text,
programming,
“Em
brained
” knowledge
CEO, chess player, fire fighter
“Em
bodied
” knowledge
Learning by doing; learning occurs not just in the brain, body memory
“En
cultured”
knowledge
Organizational culture, symbol, myth, “school spirit”
“Em
bedded”
knowledge
Moving back to tw; internship; Silicon valley, NYC; ambulance; network; embedded reporting Slide4
Culture and cognitionSlide5
Culture and cognitionSlide6
Embedded knowledge example:
Knowledge spillover
The proximity of firms within a common industry often affect how well knowledge travels among firms
The exchange of ideas among employees from different firms leads to innovations. Slide7
Knowledge
-Routinized Organizations:
Knowledge embedded in technologies, rules and procedures.
Hierarchical division of labor and control. Low skill requirement.
Example
: ‘Machine Bureaucracy’ such as a McDonalds.
Communicatio
n-Intensive Organizations:
Encultured
knowledge and collective understanding.
Communication and collaboration the key processes. Empowerment through integration.
Example: ‘Adhocracy’ such as a large management consultancyExpert-Dependent Organizations:Embodied competencies of key members.Performance of individual specialist experts is crucial. Status and power from professional reputation & qualifications.Example: ‘Professional Bureaucracy’ such as a hospital.Symbolic-Analyst-Dependent Organizations:Embrained skills of key members.Entrepreneurial problem solving. Status and power form creative achievements.Example: ‘Knowledge-intensive-firm’ such as a science-based, high tech firm.
Emphasis on collectiveendeavor
Emphasis on Contributionsof individuals
Focus on familiar problems
Focus on novel problemsSlide8
Codifiability
The ability of the firm to structure knowledge into a set of identifiable rules and relationships that can be easily communicated
Coded knowledge is
alienable
from the individual who owns the knowledge
Slide9
Decision support systemSlide10
Patent search
Codified skin care expertise
Tax preparation
P. 159. capture implicit knowledge
Examples of codification of knowledgeSlide11
Knowledge management consulting
What’s your strategy for managing knowledge HBR, March/April Slide12
Embedded knowledgeSlide13
The artifact in the work environmentSlide14
Knowledge consulting strategiesSlide15
Knowledge strategies
Codification (similar tasks)
Computer program
AI, expert system
Document
Information retrieval
Personalization (highly customized/contextualized solutions)
Knowledge sharing
People finder
Social network
National Health Service UK Health expert systemSlide16
Codifiability
The ability of the firm to structure knowledge into a set of identifiable rules and relationships that can be easily communicated
Coded knowledge is
alienable
from the individual who owns the knowledge
Slide17
Can knowledge be modeled?
Implicit knowledge
Potentially expressible
Not expressible
Expertise Slide18
Expertise
Expertise is present when an individual uses a deep structure to organize their domain knowledge, can process large chunks of information, and can work forward to solve problem in their domain of expertise ”
(Shaft, 1989)Slide19
What separate an expert from a novice
Mental models
Causal connections that govern how things work
Perceptual skills
Notice subtle cues and patterns
Sense of typicality
Something isn’t quite right…
Routines
Ways of approaching problems
Declarative knowledge
Factual information, rules, procedures Slide20
Artificial intelligence
Knowledge discovery/data mining system
Discover hidden relationships and correlations among large piles of data
Statistical methods
Expert system
Symbol manipulation/ontology
Make inferences based on known facts in the world
Representing knowledge in an ontology Slide21Slide22Slide23
YankeesSlide24
AthleticsSlide25
Policy capturing
a statistical method used in social psychology to quantify the relationship
between a person's judgment and the information that was used to make that judgment
. Policy capturing assessments rely upon regression analysis models.Slide26
Expert systemSlide27
Expert system
Every expert system consists of two principal parts: the knowledge base; and the reasoning, or inference, engine.Slide28
Knowledge base
The
knowledge base
of expert systems contains both factual and heuristic knowledge.
Commonly
agreed upon by those knowledgeable in the particular field.Slide29
Inference engine
production
rule
, or simply
rule
. A rule consists of an IF part and a THEN part
The
IF part lists a set of conditions in some logical combination.
If
the IF part of the rule is satisfied; consequently, the THEN part can be concluded,
Built in inference: if…then
Codified skin care expertiseSlide30
Procedural
Knowledge vs.
Conceptual
Knowledge Slide31
Knowledge acquisition
“Eliciting, analyzing, and interpreting the knowledge that a human expert uses when solving problems”
Knowledge bottleneck: one of the central problems faced in developing expert systems. Slide32
Interview
Self-Report
Observation
Automated-Capture
Where
in TIME:
past/ present/ future
Where
in REALISM:
real world/ simulation or scenarios
Where
in DIFFICULTIES:routine tasks/ challenging tasksWhere in GENERALITY:Abstract knowledge/ specific eventsKey attributes of CTA methodologyHow to Look:How to Look:Slide33Slide34
Subject language and ontology
Subject language
Created to represent “subjects” in
documents
The term “Butterflies” refers not to actual butterflies but a set of all indexed books about butterflies
To assist browsing and searching (human or machine)
Ontology (higher ambition)
Created to represent concepts or entities
in the world
A set or class of entities denoted by the word, such as the class consisting of all butterflies
To assist inferences made by machine Slide35
A simple semantic networkSlide36
Ontology
Created to represent
concepts or entities
in the world
A set or class of entities denoted by the word, such as the class consisting of all butterflies
To assist inferences made by machine Slide37
UMLS semantic network
Current UMLS semantic types
Current semantic network relationshipsSlide38Slide39Slide40
Wineries ontology
Red: instance
Black: class
io: is-a (e.g. subclass of, instance of )
Win
recommendationSlide41
Concept map assignment
Cognitive task analysis
Procedural knowledge
Ontology
Conceptual knowledge
Inference ?Slide42
Cognitive task analysis
Collect preliminary knowledge
Identify key tasks or concepts
Applied knowledge elicitation methods
Concept mapping Slide43
Definitions
Goals - What the
user/expert
is trying to accomplish
Operators - A (simple) action performed in service of a goal
Methods - Sequences of operators and
sub-goals
that accomplish a goal
Selection Rules - Decision points when more than one method is applicableSlide44
Task analysis: library retrieval Slide45
Concept map
Task decomposition
Build ontology
Objects
Physical
Conceptual
Attribute
Relationship
Slide46Slide47
Concept map
Concept mapping
is a technique for visualizing the relationships between different concepts.
A
concept map
is a diagram showing the relationships between concepts. Concepts are connected with labelled arrows. The relationship between concepts is articulated in linking phrases, e.g., "gives rise to", "results in", "is required by," or "contributes to". Slide48
Task analysis:
maximising the re-sale value of a car
TASKS
ACTIVITIES
FUNCTIONS
OUTPUTS
OUTCOMES
Change oil and water
Check air in tyres
Replace worn tyres
Replace headlight bulb
>
>
Clean the car
Replace faulty or worn parts
>
>
>
>
Service the car
Maintenance
Presentation
>
>
Speedometer Cable
A car that is:
Well maintained;
well presented; and
mechanically sound
Car re-sale value is maximised
Change spark plugs
Clean windows
Wash wheels
Vacuum interior
Polish paintworkSlide49
Concept map project requirements
The domain you attempt to model
Questions you asked
What you have learned from the interview process?
Did you apply the interview skills in the reading?
implicit – explicit knowledge
Expressing of the inexpressible
Story-telling
Metaphors
Comment on your map(s)
Why did you model the domain this way?
Ontology?Cognitive economy Is there alternative ways of doing it?*2 to 5 pages, not including your graphics and personal reportsSlide50
Concept mapSlide51Slide52
A concept map presentation of
cognitive task analysis Slide53
Creating a concept map
Select the domain and focus
Set up the “Parking Lot” and arrange the concepts
Begin to link the concepts
Refine the concept map
Look for new relations and cross-links
Building knowledge model Slide54
Concept map
1. Decide an area of expertise outside of your own
2. Conduct a task analysis
Decompose the task into smaller tasks and activities
Choose tasks you believe where codification strategy is more efficient
3. Conduct interview
Formalize a set of focus questions (Knowledge representation p.135)
Elicit relevant concepts from the expert
4.
Decide an ontology of the domain you are analyzing
Determine object, attribute, relationship or rule
5. Draw one or more concept maps to represent the domain Slide55
For domain overview
Could you tell me about a typical case?
Can you tell me about the last case you encountered?
Can you tell me about an usual case you heard about from same other expert?Slide56
For domain concept
Can you give me an example of X?
What is the difference between X and Y?
Does X include Y?Slide57
For domain procedures and reasoning rules
Why would you do that?
How would you do that?
What do you do at each step in this procedure?
When would you do that?
What alternatives [to the prescribed action or decision] are there?
What if it were not the case that [currently true condition]?