Examples of expert systems

Name:

Date:

Class:

Quick Questions

Shade in the A or B circle for each question.

1. Which task is a typical use of an expert system?

A Adding up the scores in a school quiz

B Diagnosing an illness from a patient's symptoms

2. A garage uses software to find why a car will not start. The software asks questions and suggests a faulty part. What is it?

A An expert system for fault finding

B A word processor

3. Which statement about mineral prospecting is correct?

A An expert system digs for the minerals itself

B An expert system can use a geologist's knowledge to suggest where minerals may be found

4. Which person does an expert system for financial advice help most?

A A customer who is not a finance expert

B Only a qualified financial adviser who already knows everything

5. Which is a good reason to use an expert system for legal advice?

A It gives access to specialist legal knowledge without needing a lawyer for simple questions

B It replaces the need for any laws to exist

Match the Pairs

Draw a line connecting each item on the left to its match on the right.

Left column		Right column
1Medical diagnosis expert system		ACar fault finding expert system
2Mineral prospecting expert system		BTakes data about rocks and land and suggests where valuable minerals or oil are likely to be found
3Fault finding in a machine needs specialist knowledge, so an expert system asks about the problem and suggests a repair		CThis describes machine learning, not an expert system. An expert system uses a fixed knowledge base and rules written from experts.
4Financial or legal advice needs specialist knowledge, so an expert system asks about the customer and gives advice		DThis is the wrong way round. Expert systems are used because the user is NOT an expert.
5An expert system is used because the user already has the specialist knowledge		EFinancial advice expert system
6A medical diagnosis expert system learns new illnesses by itself and changes its own rules		FAsks a doctor or nurse about symptoms, then suggests likely illnesses using a doctor's stored knowledge

Definitions

specialist knowledge

Fill in the Blanks

Complete each line of code by filling in the missing parts.

IF fever = TRUE ___ rash = TRUE THEN OUTPUT "Possible measles" ENDIF

IF battery = "flat" ____ OUTPUT "Charge or replace the battery" ENDIF

IF rock = "quartz" AND gold = TRUE THEN ______ "Mine this area" ENDIF

IF savings _ 1000 THEN OUTPUT "Consider investing" ENDIF

OUTPUT "Does the engine make a noise?" _____ answer

Spot the Error

Shade in the letter of the line that contains an error.

Question 1

A OUTPUT symptom

B IF symptom = "fever" THEN

C OUTPUT "Possible flu"

D ENDIF

Question 2

A INPUT noise

B IF noise = "grinding"

C OUTPUT "Check brake pads"

D ENDIF

Question 3

A INPUT rockType

B IF rockType ← "granite" THEN

C OUTPUT "Gold is likely"

D ELSE

E OUTPUT "Gold is unlikely"

F ENDIF

Question 4

A INPUT savings

B INPUT debt

C IF savings > debt THEN

D OUTPUT Consider investing

E ENDIF

Question 5

A INPUT temp

B INPUT oil

C IF temp > 90 OR oil < 20 THEN

D INPUT "Check the engine"

E ENDIF

Answer Key

Examples of expert systems

Section A: Quick Questions

B — Diagnosing an illness from a patient's symptoms

A — An expert system for fault finding

B — An expert system can use a geologist's knowledge to suggest where minerals may be found

A — A customer who is not a finance expert

A — It gives access to specialist legal knowledge without needing a lawyer for simple questions

Section B: Match the Pairs

Medical diagnosis expert system → Asks a doctor or nurse about symptoms, then suggests likely illnesses using a doctor's stored knowledge

Mineral prospecting expert system → Takes data about rocks and land and suggests where valuable minerals or oil are likely to be found

Fault finding in a machine needs specialist knowledge, so an expert system asks about the problem and suggests a repair → Car fault finding expert system

Financial or legal advice needs specialist knowledge, so an expert system asks about the customer and gives advice → Financial advice expert system

An expert system is used because the user already has the specialist knowledge → This is the wrong way round. Expert systems are used because the user is NOT an expert.

A medical diagnosis expert system learns new illnesses by itself and changes its own rules → This describes machine learning, not an expert system. An expert system uses a fixed knowledge base and rules written from experts.

Section C: Definitions

specialist knowledge — The expert knowledge that expert systems supply in areas like medical diagnosis, fault finding, mineral prospecting and financial advice, so the user does not need to be an expert.

Section D: Fill in the Blanks

IF fever = TRUE AND rash = TRUE THEN
   OUTPUT "Possible measles"
ENDIF

IF battery = "flat" THEN
   OUTPUT "Charge or replace the battery"
ENDIF

IF rock = "quartz" AND gold = TRUE THEN
   OUTPUT "Mine this area"
ENDIF

IF savings > 1000 THEN
   OUTPUT "Consider investing"
ENDIF

OUTPUT "Does the engine make a noise?"
INPUT answer

Section E: Spot the Error

A — A medical diagnosis expert system must get the symptoms from the user, who is not an expert. Line 1 should be INPUT symptom, not OUTPUT symptom.

B — An IF statement in the rule base needs THEN after the condition. Line 2 should be IF noise = "grinding" THEN.

B — The arrow ← assigns a value, but an IF needs a comparison. Line 2 should use = to compare: IF rockType = "granite" THEN.

D — Text output by the financial advice system must be in quotation marks. Line 4 should be OUTPUT "Consider investing".

D — The fault-finding system must show its advice to the user, so it uses OUTPUT. INPUT is for reading data. Line 4 should be OUTPUT "Check the engine".