cs 301 : artificial intelligence & expert...
TRANSCRIPT
![Page 1: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/1.jpg)
CS 301 : Artificial Intelligence
& Expert Systems
![Page 2: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/2.jpg)
Expert Systems
• Designed to function similar to a human expert
operating within a specific problem domain
• Used to:
– Provide an answer to a certain problem, And / or
– Clarify uncertainties where normally a human expert would be
consulted
• Often created to operate with humans, working within the
given problem domain, rather than as a replacement for them
![Page 3: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/3.jpg)
Important Components
• Knowledge Base
– Stores knowledge used by the system, usually represented in a formal logical manner
• Inference Engine
– Defines how existing knowledge may be used to derive new knowledge
![Page 4: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/4.jpg)
Knowledge Representation
• e use a si ple If … The … o se ue e elatio usi g English semantics
• ie: If [it is raining] Then [I should wear a coat]
– [it is raining] is the antecedent of the relation
– [I should wear a coat] is the consequent of the relation
• Facts can be understood as consequence relations with an
empty antecedent
– ie: If [] The [it is ai i g] is e ui ale t to the fa t that [it is ai i g]
![Page 5: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/5.jpg)
Inferring New Knowledge
k1: If [it is raining] Then [I should wear a coat]
k2: [it is raining]
result: [I should wear a coat]
• New knowledge can be constructed from existing knowledge
using inference rules
• For instance, the inference rule modus ponens can be used to
derive the consequent of a consequence relation, given that
the antecedent is true
![Page 6: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/6.jpg)
Goal Directed Reasoning
• I should wear a coat?
This ould ead easie i E glish as should I ea a oat , but we want to use the same propositional symbol as is in our knowledge base
• Inference rules are applied to knowledge base in order to achieve a particular goal
• The goal in an expert system is formed as a question, or query, to which we want the answer
![Page 7: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/7.jpg)
7
MYCIN
![Page 8: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/8.jpg)
8
History
• Thesis Project by Shortliffe @ Stanford University
• Davis, Buchanan, van Melle, and others
– Stanford Heuristic Programming Project
– Infectious Disease Group
• Project Spans a Decade
– Research started in 1972
– Original implementation completed 1976
– Resea h o ti ues i to the 80’s
![Page 9: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/9.jpg)
Tasks and Domain
• Advice for non-expert physicians with time
considerations and incomplete evidence on:
– Bacterial infections of the blood
– Expanded to meningitis and other ailments
• Disease DIAGNOSIS and Therapy SELECTION
![Page 10: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/10.jpg)
10
System Goals
• Utility
• Flexibility
• Interactive Dialogue
• Fast and Easy
![Page 11: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/11.jpg)
11
MYCIN Architecture
ConsultationSystem
ExplanationSystem
KnowledgeAcquisition
System
Q-A System
Dynamic DB
Patient DataContext Tree
Dynamic Data
Static DBRules
Parameter Properties
Context Type PropertiesTables, Lists
Physician
Expert
Physician
Expert
![Page 12: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/12.jpg)
12
Consultation System
• Performs Diagnosis and Therapy Selection
• Control Structure reads Static DB (rules) and
read/writes to Dynamic DB (patient, context)
• Linked to Explanations
• Terminal interface to Physician
![Page 13: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/13.jpg)
13
Consultation: Control Structure
1. Determine if Patient has significant infection
2. Determine likely identity of significant organisms
3. Decide which drugs are potentially useful
4. Select best drug or coverage of drugs
![Page 14: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/14.jpg)
14
ConsultationSystem
ExplanationSystem
KnowledgeAcquisition
System
Q-A System
Dynamic DB
Patient DataContext Tree
Dynamic Data
Static DBRules
Parameter Properties
Context Type PropertiesTables, Lists
Physician
Expert
Physician
Expert
![Page 15: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/15.jpg)
15
Static Database
• Rules
• Meta-Rules
• Templates
• Fed from Knowledge
Acquisition System
![Page 16: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/16.jpg)
16
Rules(Production )
• Represent Domain-specific Knowledge
• Above 450 rules in MYCIN
• Premise-Action (If-Then) Form:
• Each rule is completely modular, all relevant
context is contained in the rule.
![Page 17: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/17.jpg)
17
Meta-Rules
• Alternative to exhaustive invocation of all
rules
• Strategy rules to suggest an approach for a
given sub-goal
– Ordering rules to try first, effectively pruning the
search tree.
![Page 18: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/18.jpg)
18
Meta-Rules (continued)
• High-order Meta-Rules (i.e. Meta-Rules for
Meta-Rules)
– Powerful, but used limitedly in practice
• Impact to the Explanation System:
– (+) Encode Knowledge formerly in the Control
Structure
– (-) “o eti es eate u k e pla atio s
![Page 19: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/19.jpg)
19
Templates
• The Production Rules are all based on
Template structures
• This helps Knowledge-base expansion,
e ause the s ste a u de sta d its o representations
• Templates are updated by the system when a
new rule is entered
![Page 20: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/20.jpg)
20
ConsultationSystem
ExplanationSystem
KnowledgeAcquisition
System
Q-A System
Dynamic DB
Patient DataContext Tree
Dynamic Data
Static DBRules
Parameter Properties
Context Type PropertiesTables, Lists
Physician
Expert
Physician
Expert
![Page 21: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/21.jpg)
21
Knowledge Acquisition System
• Extends Static DB via Dialogue with
Experts
• Dialogue Driven by System
• Requires minimal training for Experts
![Page 22: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/22.jpg)
22
Knowledge Acquisition
• IF-THEN Symbolic logic was found to be easy
for experts to learn, and required little
training by the MYCIN team
• When faced with a rule, the expert must
either except it or be forced to update it
using the education process
![Page 23: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/23.jpg)
23
Knowledge reasoning
• Inexact Reasoning with Certainty Factors (CF)
• (CF are not Probability!)
• Truth of a Hypothesis is measured by a sum of
the CFs
– Premises and Rules added together
– Positive sum is confirming evidence
– Negative sum is disconfirming evidence
![Page 24: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/24.jpg)
24
ConsultationSystem
ExplanationSystem
KnowledgeAcquisition
System
Q-A System
Dynamic DB
Patient DataContext Tree
Dynamic Data
Static DBRules
Parameter Properties
Context Type PropertiesTables, Lists
Physician
Expert
Physician
Expert
![Page 25: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/25.jpg)
25
Dynamic Database
• Patient Data
• Laboratory Data
• Context Tree
• Built by Consultation System
• Used by Explanation System
![Page 26: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/26.jpg)
26
Therapy Selection
• Plan-Generate-and-Test Process
• Therapy List Creation
– Set of specific rules recommend treatments
based on the probability (not CF)
– Probabilities based on laboratory data
– One therapy rule for every organism
![Page 27: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/27.jpg)
27
Therapy Selection
• Final Selection based on:
– Sensitivity
– Contraindication Screening
– Using the minimal number of drugs and
maximizing the coverage of organisms
• Experts can ask for alternate treatments
– Therapy selection is repeated with previously
recommended drugs removed from the list
![Page 28: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/28.jpg)
28
ConsultationSystem
ExplanationSystem
KnowledgeAcquisition
System
Q-A System
Dynamic DB
Patient DataContext Tree
Dynamic Data
Static DBRules
Parameter Properties
Context Type PropertiesTables, Lists
Physician
Expert
Physician
Expert
![Page 29: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/29.jpg)
29
Explanation System
• Provides reasoning why a conclusion has
been made, or why a question is being
asked
• Q-A Module
• Reasoning Status Checker
![Page 30: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/30.jpg)
30
Preview Mechanism
• Interpreter reads rules before invoking them
• Avoids unnecessary deductive work if the sub-
goal has already been tested/determined
• Ensures self-referencing sub-goals do not
enter recursive infinite loops
![Page 31: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/31.jpg)
31
Extension works
1. Bug is uncovered, usually by Explanation
process
2. Add/Modify rules using subset of English by
experts
3. Integrating new knowledge into KB
– Found to be difficult in practice, requires
detection of contradictions, and complex
concepts become difficult to express
![Page 32: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/32.jpg)
32
MYCIN -FACTS
• Never implemented for routine clinical use
• Shown to be competent by panels of experts,
even in cases where experts themselves
disagreed on conclusions
• Key Contributions:
– Reuse of Production Rules (explanation,
knowledge acquisition models)
– Meta-Level Knowledge Use
![Page 33: CS 301 : Artificial Intelligence & Expert Systemssrmaiexpert.weebly.com/uploads/8/5/4/8/8548812/expert_system_m… · MYCIN Architecture Consultation System Explanation System Knowledge](https://reader030.vdocuments.us/reader030/viewer/2022033121/5eb800879a2f8105be7cbe93/html5/thumbnails/33.jpg)
33
References
• Jason Walonoski, Standsford University.
• Davis, Buchanan, Shortliffe. Production Rules as a Representation for a Knowledge-Based Consultation System. Artificial Intelligence, 1979.
• William van Melle. The Structure of the MYCIN System. International Journal of Man-Machine Studies, 1978.
• Shortliffe. Details of the Consultation System. Computer-
Based Medical Consultations: MYCIN, 1976.
• Jadzia Cendrowska, Max Bramer. Chapter 15?
• Major Lesso s Fro this Work • William J. Clancey. Details of the Revised Therapy Algorithm.
1977