Scribd
Upload
82 views2 pages

WWW Igcseict Info Theory 7 - 2 Expert Index HTML

The document discusses expert systems, which are computer programs that aim to mimic human experts. It describes the components of expert systems, including the user interface, knowledge base, and inference engine. It also provides examples of where expert systems are used, such as for medical diagnosis, games, financial advice, and more. Potential issues with expert systems are also covered.

Uploaded by

Hitesh Bhavnani
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
82 views2 pages

WWW Igcseict Info Theory 7 - 2 Expert Index HTML

The document discusses expert systems, which are computer programs that aim to mimic human experts. It describes the components of expert systems, including the user interface, knowledge base, and inference engine. It also provides examples of where expert systems are used, such as for medical diagnosis, games, financial advice, and more. Potential issues with expert systems are also covered.

Uploaded by

Hitesh Bhavnani
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
You are on page 1/ 2

Expert 

Systems | IGCSE ICT
You are here →  Home  →  Theory Notes  →  7.2 ICT Use in the Workplace  →  Expert Systems

What is an Expert System?

An  expert  system  is  computer  software  that  attempts  to  act  like  a
human expert on a particular subject area. 

Expert  systems  are  often  used  to  advise  non­experts  in  situations
where  a  human  expert  in  unavailable  (for  example  it  may  be  too
expensive to employ a human expert, or it might be a difficult to reach
location).

How Do Expert Systems Work?

An expert system is made up of three parts:

A  user interface  ­ This  is  the  system  that  allows  a  non­expert


user  to  query  (question)  the  expert  system,  and  to  receive
advice. The user­interface is designed to be a simple to use as
possible.
A knowledge base ­ This is a collection of facts and rules. The
knowledge  base  is  created  from  information  provided  by
human experts
An  inference  engine  ­  This  acts  rather  like  a  search  engine,
examining the knowledge base for information that matches the
user's query

The  non­expert  user  queries  the  expert  system.  This  is  done  by
asking a question, or by answering questions asked by the expert
system.

The inference engine uses the query to search the knowledge base
and then provides an answer or some advice to the user.

Where Are Expert Systems Used?

Medical  diagnosis  (the  knowledge  base  would  contain  medical


information, the symptoms of the patient would be used as the query,
and the advice would be a diagnose of the patient’s illness) 

Playing  strategy  games  like  chess  against  a  computer  (the


knowledge  base  would  contain  strategies  and  moves,  the  player's
moves  would  be  used  as  the  query,  and  the  output  would  be  the
computer's 'expert' moves)

Providing financial advice ­ whether to invest in a business, etc. (the
knowledge  base  would  contain  data  about  the  performance  of
financial markets and businesses in the past)
You can try a demonstration of a
Helping to identify items such  as  plants  /  animals  /  rocks  /  etc.  (the car fault diagnosis system on­line
knowledge  base  would  contain  characteristics  of  every  item,  the here .
details  of  an  unknown  item  would  be  used  as  the  query,  and  the
advice would be a likely identification)

Helping to discover locations to drill for water / oil (the  knowledge
base would contain characteristics of likely rock formations where oil /
water  could  be  found,  the  details  of  a  particular  location  would  be
used as the query, and the advice would be the likelihood of finding oil
/ water there)

Helping to diagnose car engine problems (like medical diagnosis, but
for cars!)
The Scottish health service has a
self­diagnosis expert system you
Can Expert Systems Make Mistakes? can try on­line  here .

Human experts make mistakes all the time (people forget things, etc.)
so you might imagine that a computer­based expert system would be
much better to have around.

However expert systems can some problems:

Can't  easily  adapt  to  new  circumstances  (e.g.  if  they  are
presented  with  totally  unexpected  data,  they  are  unable  to
process it)
Can be difficult to use (if the non­expert user makes mistakes
when  using  the  system,  the  resulting  advice  could  be  very
wrong)
They have no 'common sense' (a human user tends to notice
obvious errors, whereas a computer wouldn't)

Next Up → Manufacturing Products

Sorry to ask you this, but...
This site took quite a bit of time and effort to create, and it costs me money to
 
keep it up and running. If the site has helped you, please consider showing
your appreciation by donating a little towards the site's running costs.

Thank you!
 
Sponsored Links...

You are here →  Home  →  Theory Notes  →  7.2 ICT Use in the Workplace  →  Expert Systems

igcseict.info © Steve Copley

You might also like