2. Introduction
a method by which the knowledge engineer acquires knowledge from the
expert.
a record or documentation of the expert’s step-by-step information-
processing and decision-making behaviour.
Later transcribed for further analysis called protocol and coded by knowledge
Engineer.
Particularly useful in analyzing dynamic reasoning behavior
3. Assumption
To produce a knowledge-level representation of the problem solving process
in terms of task level framework
A structured procedure based on the theoretical framework ensuring reliable
encoding
A method is used to achieve consensus with systematic study
4. Introduction
Analysis of the expert actually solving problems in the domain
Online(Concurrent) protocol analysis
Self-report (also called think-alound)
Shadowing
Offline protocol analysis
Retrospective verbalization of the problem-solving
5. Process of Protocol Analysis
Provide the expert with a full range of information normally associated with a
task.
Ask the expert to verbalize the task
Make statements by transcribing the verbal protocols.
Gather the statements that seem to have high information content.
Simplify and rewrite the collected statements and construct a table of
production rules from the collected statements.
Produce a series of models by using the production rules
6. Phases
Phase [I] Knowledge Guided Behavior[KGB]
Phase[II] Consensus In Analysis[CIA] among different knowledge engineers
Phase[III] Builds New Descriptions[BND] in terms of problem solving language
7. Sample Question and Answer
E: you said “it sometimes happens”, “it jumps”. Now does it jump towards the periphery or towards
the centre?
I: I think it exclusively jump towards the periphery, no, towards centre, excuse me.
E: towards the centre. Well before it was jumping the other way- the same part of a piece was
played over and over.
I: yes but then it jumps….
E: it is playing from the periphery towards the centre, so when it jumps toward periphery it
repeats.
and so on….
8. Example
Expert explanation(Protocol):”As we can see the Porche 911 is a very
comfortable and fast car, so it is very expensive”.
Step I:Divide the text into segments:
Line Segments
1 As we can see
2 The Porsche 911
3 is very comfortable
4 and fast car
5 so it is
6 expensive
9. Example cont…
Step II:
Identify concepts, characteristics, value,
relation and operators.
Element Id
As we can see Relation
The Porsche 911 Concept
is very comfortable Value
and fast car Value
so it is Operator
expensive Concept
Step III:
Specify the relation
Concepts Characteristics Values
Porsche 911 (Comfort) Very Comfortable
(Speed) fast
Expensive
10. Example cont…
Step IV:Draw Search identification
so it is
Facts to get this state:
-fast
-very comfortable
Step V:After performing search identification, Write the rule associated to it:
Rule “It’s an expensive car”
when
$porsche: Porsche911(speed==“fast”, comfort==“very comfortable”)
then
System.out.println(“Car:” $porsche.getname()+” is an expensive car”);
end
What kind of
car is this? Expensive
11. Areas of Protocol Analysis
Crypto arithmetic(Newell and SImon)
Medical Diagnosis(Myers, Fox, Pegram and Greaves)
Design of mechanical component (Lewis)
12. Areas of Protocol Analysis
Crypto arithmetic
Medical Diagnosis
Design of mechanical component
13. Advantages
The expert consciously considers decision-making heuristics.
The expert consciously considers decision alternatives, attributes, and values.
The knowledge engineer can observe and analyze decision-making behavior.
The knowledge engineer can record, and later analyze with the expert, key
decision points.
14. Drawbacks
The expert must be aware of why he or she makes a decision.
The expert must be able to categorize major decision alternatives.
The expert must be able to verbalize the attributes and values of a decision
alternative.
The expert must be able to reason about the selection of a given alternative.
A view of decision making is subjective. Explanations may not track with
reasoning.
15. Summary
1. Provide the expert with a full range of information normally associated
with a task.
2. Ask the expert to verbalize the task in the same manner as would be done
normally while verbalizing his or her decision process and record the
verbalization on tape.
3. Make statements by transcribing the verbal protocols.
4. Gather the statements that seem to have high information content.
5. Simplify and rewrite the collected statements and construct a table of
production rules from the collected statements.
6. Produce a series of models by using the production rules