You are on page 1of 5

INSTITUTE OF COMPUTER SCIENCE AND DIGITAL INNOVATION ONLINE

FINAL ASSESSMENT 

Student ID (in Figures) :

1 0 0 1 8 5 2 4 3 1

Student ID (in Words) : One Zero Zero One Eight Five Two Four Three One
Course Code & Name : CC306 Intelligent System
Semester : January 2021
Lecturer : Monica Regmy
Duration : 17th February 20211, 8:00 am – 9:30 pm 

INSTRUCTIONS TO CANDIDATES: 
Answer All Questions [Total: 100 marks] 

This question paper consists of 4 pages, excluding the cover page. Candidates 
are required to answer the questions and submit answers (softcopy) on CN 
Dropbox by the deadline given.

This question paper must be submitted along with all used and/or unused rough  papers and/or
graph papers (if any). Candidates are NOT allowed to take any  examination materials out of the
examination hall. 

Warning:  
The University Examination Board of UCSI University regards cheating as a most  serious
offence and will not hesitate to mete out the appropriate punitive actions  according to the
severity of the offence committed, and in accordance with the clauses  stipulated in the Students’
Handbook, up to and including expulsion from UCSI University.
1. Artificial Intelligence (AI) is getting more popular nowadays.
a) Do you think a computer can talk or understand speech? Can computer learn and do
decision-making as well? Explain your opinion. (8 marks)
b) Describe a Turing test for AI and justify its validity from your modern standpoint. (5
marks)

2. The objective of knowledge representation is to express knowledge in computer tractable


form. Convert the following knowledge into a semantic network. (13 marks)

Tom is a cat. Tom caught a bird.


Tom is owned by John. Tom in ginger in color.
A cat is a mammal.
A bird is an animal.
All mammals are animals.
Mammals have fur.

Animal

is-a is-a
Mammal
Bird
has fur

is-a

Cat

is-a

Tom

Catches bird, is ginger


colored

owns

John
3. Based on your knowledge, write FIVE (5) rules that you can extract from the following
conversation. (5 marks)
• Patient: I have a headache
• System: I suggest you take a painkiller. Do you have a sore throat?
• Patient: Yes, I do.
• System: Have you been coughing?
• Patient: No, why do you ask?
• System: If you had a headache and a sore throat and had been coughing. I would have to
conclude you might have the flu. It doesn’t sound as though you have the flu? Have you
been feeling tired recently?
• Patient: Yes, I have.
• System: It sounds as though you may have glandular fever or tonsillitis.

Ans: Five rules inferred from the conversation above are listed below:

1. If you have a headache, then take a painkiller.

2. If you have a headache


and a sore throat
then avoid painkiller

3. If you have a headache


and a sore throat
and have been coughing
and distorted sound
then you have a flu

4. If you have a headache


and a sore throat
and have been coughing
and feeling tired lately
then you have glandular fever

5. If you have a headache


and a sore throat
and have been coughing
and feeling tired lately
then you have tonsillitis
b) Draw an inference chain based on the rules extracted in question 3 (a). (5 marks)
Ans:

Eat painkiller

Have a headache

Have a sore throat

Flu Glandular/
Tonsillitis

Coughing

Distorted sound

Feeling tired

c) Based on your opinion, which type of chaining is better for medical expert system? Give reasons
to support your opinion. (4 marks)
Ans: In the field of medicine, there exists information on patient that needs to be provided first to
begin the treatment. These information can include: age, symptoms, sex, etc that determines the
choice of medicine as well. Moreover, the combination of common symptoms makes it even
difficult to pin point at a confident solution.

Applying backward chaining would require to define the goal first, before the inference engine
attempts to provide possible solutions. Hence, backward chaining is incompatible for medical
expert system as we require a remedy inferred from the patient data.

Forward chaining is a data-driven approach to reach a suitable conclusion(s). The preconditions of a


patient can serve as the data and the knowledge of a doctor can serve as the rule base for the engine.
This approach is greatly efficient in the sense forward-chaining can infer multiple facts from the
existing rule base and provide more confident set of possible treatments.
4. a) Briefly describe rules in a rule-based expert system. (7 marks)
Ans: Expert system portrays the expertise of an experienced personnel who have extensive
knowledge about a particular domain(s). This knowledge exists in the form of facts and
rules in the expert system.

• Rules in an expert system is a problem solving technique to represent knowledge in a


IF...THEN structure. For instance: IF it is raining, THEN you should carry an umbrella.

• IF part of a rule is termed antecedent. It holds the condition or the prerequisite that must
be fulfilled before acting on conclusion defined in THEN part. The consequent
(THEN) holds the action to be performed when the condition in IF is satisfied.

• Logical operator can be used to add multiple conditions as below:


◦ IF <antecedent 1> AND <antecedent 2> AND <antecedent 3> THEN <consequent>
◦ IF <antecedent 1> OR <antecedent 2> OR <antecedent 3> THEN <consequent>

• The two parts that makes up an antecedent are: object and a value.
For instance: IF age of the participant < 20
AND +2 passed(years) < 3
THEN candidate can apply

• The two parts that makes up an antecedent are: object and a value.
For instance: IF age of the participant < 20 AND +2 passed(years) < 3 THEN candidate
can apply.

b) Explain the tasks of domain expert and knowledge engineer in the expert system
development team. (3 marks)
Ans: A domain expert consistently communicates with the project manager and knowledge
engineer to pour his/her knowledge and expertise that will go into the final expert system.
They hold the responsibility to help other players conceptualize the facts and rules to
understand the system before implementation.

The knowledge feed by the domain expert is analyzed and brought into practical terms by
the knowledge engineer. They are responsible for designing, building, testing and deploying
the expert system. Selection of reasoning method, knowledge representation method and
handling facts and rules are some common tasks of a knowledge engineer.

You might also like