0% found this document useful (0 votes)
6 views87 pages

Chapter 6

The document discusses the concepts of artificial intelligence (AI) and expert systems, highlighting their structures, advantages, and disadvantages. It differentiates between expert systems, which are rule-based, and machine learning systems that utilize statistical modeling. Examples of expert systems in various fields, such as medical diagnosis and route scheduling, are also provided.
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
6 views87 pages

Chapter 6

The document discusses the concepts of artificial intelligence (AI) and expert systems, highlighting their structures, advantages, and disadvantages. It differentiates between expert systems, which are rule-based, and machine learning systems that utilize statistical modeling. Examples of expert systems in various fields, such as medical diagnosis and route scheduling, are also provided.
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd



• •

• •
the overhead sometimes be incorrect, for
weather conditions example if snow
is fed constantly to is forecast and does
air traffic controllers. not arrive.
that can be carried out
continuously over many days. Sensors detect changes in
environmental conditions, microprocessors can calculate
required alterations to the experimental settings and actuators
can turn on heaters, coolers or water supplies as required.




Based on the definition of a
robot, is chat-bot a robot?

No, because chat-bot does
not have mechanical
structure.






• •

• •








E X A M Q U ES T I O N ( M A RC H 1 9 )
E X A M Q U ES T I O N ( M A RC H 1 9 )
Chapter 6.3
AI DEFINITION
A branch of computer science dealing with the simulation of INTELLIGENT

human behaviour (reasoning, speech, sight) by a computer.

WHAT IS INTELLIGENCE?
The thought process that you might have ...
ARTIFICIAL
INTELLIGENCE

CATEGORY 2:
MACHINE
LEARNING

CATEGORY 1:
EXPERT
SYSTEM
ARTIFICIAL
INTELLIGENCE

CATEGORY 1:
EXPERT
SYSTEM
EXPERT SYSTEM
EXPERT SYSTEM
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE

USER INTERFACE

INTERACT
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE KNOWLEDGE BASE

INTERACT
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE KNOWLEDGE BASE

INTERACT

RULE BASE
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE KNOWLEDGE BASE

INTERACT

INFERENCE BASE
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE KNOWLEDGE BASE

INTERACT

INFERENCE ENGINE RULE BASE


STRUCTURE OF AN EXPERT SYSTEM
USER EXPLANATION
INTERFACE SYSTEM KNOWLEDGE BASE

INTERACT •

INFERENCE ENGINE RULE BASE


STRUCTURE OF AN EXPERT SYSTEM
USER EXPLANATION
INTERFACE SYSTEM KNOWLEDGE BASE

INTERACT

INFERENCE ENGINE RULE BASE


EXAMPLE(1) OF AN EXPERT SYSTEM

MEDICAL DIAGNOSIS
USER EXPLANATION
INTERFACE SYSTEM KNOWLEDGE BASE

SUGGEST LIBRARY OF
POSSIBLE SYMPTOM
DISEASE AND RELATIONS TO
ACTION STEPS DISEASE

INTERACT

IF USER HAS
MULTIPLE
SYMPTOMS,
SUGGEST
DISEASE
INFERENCE ENGINE RULE BASE
EXAMPLE(2) OF AN EXPERT SYSTEM
ROUTE SCHEDULING FOR DELIVERY SERVICE
USER EXPLANATION
INTERFACE SYSTEM KNOWLEDGE BASE

SUGGEST THE NUMBER OF ROUTES,


BEST ROUTE TO PARCELS AND DELIVERY
BE TAKEN POINTS

INTERACT

IF TRAFFIC JAM,
DON'T PICK THAT
ROUTE

INFERENCE ENGINE RULE BASE


EXAMPLE(3) OF AN EXPERT SYSTEM
DIAGNOSIS OF BREAST CANCER
USER EXPLANATION
INTERFACE SYSTEM KNOWLEDGE BASE

PREDICT PROBABILITY OF MAMMOGRAPHYS


A PERSON GETTING WITH ALL TYPES OF
BREAST CANCER CANCERS / NORMAL
MAMMOGRAPHY

INTERACT

IF HUGEWHITE SPOTS
ARE OBSERVED,
CATEGORISE AS CANCEL

INFERENCE ENGINE RULE BASE


EXAMPLE(3) OF AN EXPERT SYSTEM
DIAGNOSIS OF BREAST CANCER
KNOWLEDGE BASE

MAMMOGRAPHYS
WITH ALL TYPES OF
CANCERS / NORMAL
MAMMOGRAPHY
OTHER EXAMPLES OF AN EXPERT SYSTEM

• MINERAL PROSPECTING
• PLANT AND ANIMAL IDENTIFICATION
• FINANCIAL PLANNING
• CAR ENGINE FAULT SYSTEMS
• BANK LOAN APPROVAL SYSTEM
EXPERT SYSTEM
Advantages
1. HIGH LEVEL OF EXPERTISE
2. HIGH ACCURACY
3. CONSISTENT RESULT
4. HAVE THE ABILITY TO STORE VAST AMOUNTS OF IDEAS AND FACTS
5. THEY HAVE VERY FAST RESPONSE TIME

Disadvantages
1. SET UP AND MAINTENANCE COSTS ARE VERY HIGH
2. THEY ARE ONLY AS GOOD AS THE INFORMATION ENTERED INTO THE SYSTEM
ARTIFICIAL
INTELLIGENCE

CATEGORY 2:
MACHINE
LEARNING
MACHINE LEARNING
MACHINE LEARNING
EXAMPLE (1)
SPAM FILTER

"CONGRATULATION, YOU HAVE WON 2 MILLIONS


DOLLAR $$$ IN THE NATIONAL LOTTERY. SIGN UP
AS OUR MEMBER NOW TO REDEEM YOUR PRIZE"

"HOW DO WE TEACH THE MACHINE THAT THIS


EMAIL IS INDEED A SPAM EMAIL?"
PROCESS

A MACHINE
LEARNING
ALGORITHM
COLLECTS DATA
ABOUT EMAILS
PROCESS "CONGRATULATION, YOU
HAVE WON 2 MILLIONS
DOLLAR $$$ IN THE
NATIONAL LOTTERY. SIGN UP
AS OUR MEMBER NOW TO
REDEEM YOUR PRIZE"

A MACHINE CARRIES OUT A


LEARNING CLEANING
ALGORITHM PROCESS TO "CONGRATULATION, WON 2
COLLECTS DATA REMOVE STOP MILLIONS DOLLAR $$$
ABOUT EMAILS WORDS (THE, NATIONAL LOTTERY SIGN UP
OUR MEMBER NOW
AN, A)
REDEEM PRIZE"
PROCESS

A MACHINE CARRIES OUT A CERTAIN WORDS/ PHRASES ARE FREQUENTLY


LEARNING CLEANING USED IN EMAIL THAT COULD INDICATE THAT
ALGORITHM PROCESS TO THE INCOMING EMAIL IS VERY LIKELY TO BE
COLLECTS DATA REMOVE STOP SPAM
ABOUT EMAILS WORDS (THE, (LOTTERY, EARN, FULL-REFUND).
AN, A)
PROCESS THE MACHINE LEARNING MODEL IS BUILT
AND A TRAINING DATA SET IS USED TO
TRAIN THE MODEL AND MAKE IT USING
PAST EMAILS KNOWN THE BE SPAM. ONCE
IT IS EVALUATED, THE MODEL IS FINE-
TUNED AND TESTED LIVE

A MACHINE CARRIES OUT A CERTAIN WORDS/ PHRASES ARE FREQUENTLY


LEARNING CLEANING USED IN EMAIL THAT COULD INDICATE THAT
ALGORITHM PROCESS TO THE INCOMING EMAIL IS VERY LIKELY TO BE
COLLECTS DATA REMOVE STOP SPAM
ABOUT EMAILS WORDS (THE, (LOTTERY, EARN, FULL-REFUND).
AN, A)
EXAMPLE (2)
AUTO TAG - PHOTOGRAPH
DIFFERENCE BETWEEN THE EXPERT SYSTEM AND
MACHINE LEARNING

CATEGORY 1: CATEGORY 2:
EXPERT THE MAIN DIFFERENCE IS THAT EXPERT
MACHINE
SYSTEMS ARE RULE BASED SYSTEMS WHILE
SYSTEM MODERN MACHINE LEARNING (ML) ARE LEARNING
BASED ON STATISTICAL MODELING OF DATA.
THAT IS, AN EXPERT SYSTEM USES IF-THEN
STATEMENTS WHEN DOING INFERENCE
WHILE AN ML SYSTEM PROJECTS THE INPUT
INTO SOME MODEL SPACE.

You might also like