03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
•
•
•
03 Af
34 sh
54 an
97
14
9
•
•
03 Af
34 sh
54 an
•
•
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
9
54 an
14
34 sh
97
the overhead sometimes be incorrect, for
03 Af
weather conditions example if snow
is fed constantly to is forecast and does
air traffic controllers. not arrive.
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
that can be carried out
9
continuously over many days. Sensors detect changes in
54 an
14
environmental conditions, microprocessors can calculate
34 sh
97
required alterations to the experimental settings and actuators
03 Af
can turn on heaters, coolers or water supplies as required.
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
•
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
•
•
34 sh
54 an
97
14
9
9
54 an
14
•
34 sh
Based on the definition of a
97
03 Af
robot, is chat-bot a robot?
9
54 an
14
•
34 sh
No, because chat-bot does
97
03 Af
not have mechanical
structure.
03 Af
34 sh
54 an
97
14
9
•
•
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
•
•
•
•
97
14
9
•
•
•
03 Af
34 sh
54 an
97
14
9
•
•
03 Af
34 sh
54 an
•
•
97
14
9
•
•
•
•
•
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
•
•
•
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
•
•
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
03 Af
34 sh
54 an
97
14
9
E X A M Q U ES T I O N ( M A RC H 1 9 )
9
54 an
14
34 sh
97
03 Af
E X A M Q U ES T I O N ( M A RC H 1 9 )
9
54 an
14
34 sh
97
03 Af
03 Af
34 sh
54 an
97
14
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.
9
54 an
14
34 sh
97
WHAT IS INTELLIGENCE?
03 Af
ARTIFICIAL
INTELLIGENCE
CATEGORY 2:
9
54 an
14
MACHINE
34 sh
97
LEARNING
03 Af
CATEGORY 1:
EXPERT
SYSTEM
ARTIFICIAL
INTELLIGENCE
9
54 an
14
34 sh
97
03 Af
CATEGORY 1:
EXPERT
SYSTEM
EXPERT SYSTEM
03 Af
34 sh
54 an
97
14
9
EXPERT SYSTEM
03 Af
34 sh
54 an
97
14
9
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE
9
USER INTERFACE
54 an
14
34 sh
97
03 Af
INTERACT
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE KNOWLEDGE BASE
9
54 an
14
34 sh
97
03 Af
INTERACT
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE KNOWLEDGE BASE
9
54 an
14
34 sh
97
03 Af
INTERACT
RULE BASE
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE KNOWLEDGE BASE
9
54 an
14
34 sh
97
03 Af
INTERACT
INFERENCE BASE
STRUCTURE OF AN EXPERT SYSTEM
USER
INTERFACE KNOWLEDGE BASE
9
54 an
14
34 sh
97
•
03 Af
INTERACT
INFERENCE ENGINE RULE BASE
STRUCTURE OF AN EXPERT SYSTEM
USER EXPLANATION
INTERFACE SYSTEM KNOWLEDGE BASE
9
54 an
14
34 sh
97
•
03 Af
INTERACT •
INFERENCE ENGINE RULE BASE
STRUCTURE OF AN EXPERT SYSTEM
USER EXPLANATION
INTERFACE SYSTEM KNOWLEDGE BASE
9
54 an
14
34 sh
97
03 Af
INTERACT
INFERENCE ENGINE RULE BASE
EXAMPLE(1) OF AN EXPERT SYSTEM
MEDICAL DIAGNOSIS
USER EXPLANATION
INTERFACE SYSTEM KNOWLEDGE BASE
9
54 an
SUGGEST LIBRARY OF
14
POSSIBLE SYMPTOM
34 sh
97
DISEASE AND RELATIONS TO
ACTION STEPS DISEASE
03 Af
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
9
54 an
SUGGEST THE NUMBER OF ROUTES,
14
BEST ROUTE TO PARCELS AND DELIVERY
34 sh
97
BE TAKEN POINTS
03 Af
INTERACT
IF TRAFFIC JAM,
DON'T PICK THAT
ROUTE
INFERENCE ENGINE RULE BASE
OTHER EXAMPLES OF AN EXPERT SYSTEM
• MINERAL PROSPECTING
• PLANT AND ANIMAL IDENTIFICATION
9
54 an
14
• FINANCIAL PLANNING
34 sh
97
• CAR ENGINE FAULT SYSTEMS
03 Af
• BANK LOAN APPROVAL SYSTEM
EXPERT SYSTEM
Advantages
1. HIGH LEVEL OF EXPERTISE
2. HIGH ACCURACY
9
54 an
14
3. CONSISTENT RESULT
4. HAVE THE ABILITY TO STORE VAST AMOUNTS OF IDEAS AND FACTS
34 sh
97
5. THEY HAVE VERY FAST RESPONSE TIME
03 Af
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:
9
54 an
14
MACHINE
34 sh
97
LEARNING
03 Af
MACHINE LEARNING
03 Af
34 sh
54 an
97
14
9
MACHINE LEARNING
03 Af
34 sh
54 an
97
14
9
EXAMPLE (1)
SPAM FILTER
9
54 an
14
"CONGRATULATION, YOU HAVE WON 2 MILLIONS
34 sh
97
DOLLAR $$$ IN THE NATIONAL LOTTERY. SIGN UP
03 Af
AS OUR MEMBER NOW TO REDEEM YOUR PRIZE"
"HOW DO WE TEACH THE MACHINE THAT THIS
EMAIL IS INDEED A SPAM EMAIL?"
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
9
IT IS EVALUATED, THE MODEL IS FINE-
54 an
14
TUNED AND TESTED LIVE
34 sh
97
03 Af
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
9
54 an
14
34 sh
97
03 Af
DIFFERENCE BETWEEN THE EXPERT SYSTEM AND
MACHINE LEARNING
CATEGORY 1: CATEGORY 2:
9
EXPERT THE MAIN DIFFERENCE IS THAT EXPERT
MACHINE
54 an
14
SYSTEMS ARE RULE BASED SYSTEMS WHILE
SYSTEM MODERN MACHINE LEARNING (ML) ARE LEARNING
34 sh
97
BASED ON STATISTICAL MODELING OF DATA.
THAT IS, AN EXPERT SYSTEM USES IF-THEN
03 Af
STATEMENTS WHEN DOING INFERENCE
WHILE AN ML SYSTEM PROJECTS THE INPUT
INTO SOME MODEL SPACE.
