Professional Documents
Culture Documents
Informal Introduction to
Similarity-Based and
Case-Based Reasoning
Stand 20.12.00
Recommended References
This lecture is not intended to provide a complete
introduction into case based reasoning. It will rather deal
only with those aspects which are used in applications to
e-c. For additional readings we recommend:
A. Aamodt, E. Plaza: Case-Based Reasoning: Foundational
Issues, Methodological Variations, and System Approaches. AI
Communications 7(1) (1994), S.39-59.
M. Lenz, B. Bartsch-Sprl, H.-D. Burkhard, S. Wess (eds.): CaseBased Reasoning Technology. Springer Lecture Notes in AI 1400,
1998.
R. Bergmann, S. Breen, M. Gker, M. Manago, S. Wess:
Developing Industrial Case-Based Reasoning Application - The
INRECA- Methodology. Springer Lecture Notes in AI 1612, 1999.
-2-
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
-4-
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
What is CBR?
-5-
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
-6-
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
-7-
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Extensions to E-Commerce
The extensions to applications in electronic commerce have
been developed in various research projects like
-8-
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
-9-
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Mathematically
Algorithmically
With respect to software technology
Supported by experiments and applications
Business success
- 10 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
What is a Case ?
A case has two parts:
Description of a problem or a set of problems
(generalized case)
Description of the solution of this problem
(formally or informally)
Possibly additions like explanations, comments on the quality
of the solution etc.
Cases represent experiences :
They record how a problem was solved in the past
- 11 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 12 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Object-oriented representations
Case: collection of objects (instances of classes)
Required for complex and structured objects
- 13 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Problem
Solution ?
Solution
adaptation
Solution
of the case
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 15 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Vocabulay
Solution
transformation
Case
base
Storage
Compilation
Data
Information
Knowledge
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
diagnosis
classification
repair
Domain rules
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Case-Based Diagnosis:
a case describes a diagnostic situation and contains:
description of the symptoms
description of the failure and the cause
description of a repair strategy
store a collection of cases in a case base
find case similar to current problem and reuse repair strategy
- 18 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
C
A
S
E
1
Value
Problem (Symptoms)
Problem: Front light doesnt work
Car: VW Golf IV, 1.6 l
Year: 1998
Battery voltage: 13,6 V
State of lights: OK
State of light switch: OK
Solution
Diagnosis: Front light fuse defect
Repair: Replace front light fuse
- 19 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Each case
describes one
particular situation
All cases are
independent of
each other
C
A
S
E
1
C
A
S
E
2
Problem (Symptoms)
Year: 1996
State of lights: OK
Car: Audi A4
Year: 1997
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Value
Problem (Symptom):
Problem: Break light doesnt work
Car: Audi 80
Year: 1989
Battery voltage: 12.6 V
State of light: OK
- 21 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Similar?
Similar?
C
A
S
E
x
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Very similar
Examples:
Feature: Problem
0.8
0.4
12.6 V
0.9
13.6 V
0.1
6.7 V
12.6 V
Different features have different importance (weights)!
High importance: Problem, Battery voltage, State of light, ...
Low importance: Car, Year, ...
- 23 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
0.8
0.4
0.6
0.9
1.0
Problem (Symptoms)
Problem: Front light doesnt work
Car: VW Golf III, 1.6 l
Year: 1996
Battery voltage: 13.6 V
State of lights: OK
State of light switch: OK
Solution
Diagnosis: Front light fuse defect
Repair: Replace front light fuse
- 24 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
0.8
0.8
0.4
0.95
0
Problem (Symptoms)
Problem: Front light doesnt work
Car: Audi A4
Year: 1997
Battery voltage: 12.9 V
State of lights: surface damaged
State of light switch: OK
Solution
Diagnosis: Front light fuse defect
Repair: Replace front light fuse
- 25 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Problem (Symptoms):
Problem: Front light doesnt work
...
Solution:
Diagnosis: Front light fuse defect
Repair: Replace front light fuse
Problem (Symptom):
Problem: Break light doesnt work
Car: Audi 80
Year: 1989
Battery voltage: 12,6 V
State of light: OK
Adapt Solution:
How do differences in the
problem affect the solution?
New Solution:
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
C
A
S
E
3
Problem (Symptoms):
Problem: Break light doesnt work
Car: Audi 80
Year: 1989
Battery voltage: 12.6 V
State of lights: OK
State of light switch: OK
Solution:
Diagnosis: break light fuse defect
Repair: replace break light fuse
- 27 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Customer Classification
In e-c another classification problem arises: To define
classes of customers with the same behavior and
treat the customers according to their class:
Observations
about the
customer
Customer
class
Customer
treatment
Domain
knowledge
Inexact
classification
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Retrieve:
Problem
New
Case
Re
tai
n
Learned
Case
Confirmed
Solution
Case
Base
Reuse:
New
Retrieved Case
Case
Knowledge
Re
vis
e
Tested/
Repaired
Case
Re
t r ie
ve
Re
us
e
Solved
Case
Revise:
Evaluate the applicability of the
proposed solution in the real-world.
Retain:
Suggested
Solution
- 29 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Similarities (1)
Similarities are described by measures with
numerical values
They operate on
problem descriptions, demands, products ,...
Intention:
The more similar two problem descriptions C and D are,
the more useful it is two use one of the solutions also
for the other problem.
The more similar a demand and a product are the more
useful is the product for satisfying the demand.
- 31 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Similarities (2)
Given a fixed problem (demand) C
A similarity measure introduces a partial ordering (to be
more or less similar to C) on
the set of problems and therefore also on the case base
the set of products and therefore on the the product base.
- 32 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Similarities (3)
An important consequence of the intention is that
similarity is strongly related to utility
Utility is provided by the customer and should reflect
his interests and needs
Similarity is secondary because it is used to find
solutions for the customer.
Similarity is therefore not an absolute notion like truth
but a problem dependent notion.
- 33 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Similarities (4)
The similarity measure is the central element to
navigate through the space of possible solutions or
possible products.
Instead of presenting the exact solution similarity is a
concept to approximates it.
Even when the exact or optimal solution is not
available or too difficult to achieve one comes still up
with at least a suggestion for the solution.
- 34 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 35 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
This partitions this case base (for the actual problem into three
parts (so-called rough sets): accepted cases, unaccepted cases
and an uncertainty set.
The same works for product bases.
- 37 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 39 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 40 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
No revise phase
Verification of the solution by computer simulation
Verification / evaluation of the solution in the real world
For products: Technical or customer evaluation, buying decision
- 41 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Methods
- 42 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
CBR Systems
Solution
Knowledge Acquisition
Domain
Knowledge
!!
KBS
Knowledge Base
Problem
Acquisition of general
knowledge is very difficult !!
- 44 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
CBR System
- 46 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
CBR System
- 47 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Knowledge,
Models
New
Case
RETRIEVE
RETAIN
Verified
Solution
Case Base
Knowledge Model
REVISE
Adapted
Solution
Increasing
Knowledge Centralization
Expert Systems
Data,
Examples
REUSE
Open Database
Increasing
Example Orientation
CBR Systems
- 48 -
Similar
Case
Databases
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Search Engine
CBR
- 49 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
CBR Systems
search is processed by
keywords
- 50 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 51 -
Database
Expert
System
CBR
Information
Retrieval
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Initial
demands
Re
t r ie
ve
Re
fin
User
demands
Product
Base
Evaluation
Reuse:
User
Retrieved
demands
product
Knowledge
Re
vis
e
Tested
product
Modified
product
Re
us
e
Revise:
Evaluate the product in the real world.
Refine:
Learn from customer behavior
Offer
- 52 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 53 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
stored
product ?
(class ?)
reuse
- 54 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 55 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 56 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 57 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
- 58 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern
Summary
Reuse,
Revise,
Retain
- 59 -
(c) 2000 Dr. Ralph Bergmann and Prof. Dr. Michael M. Richter, Universitt Kaiserslautern