Applications Of AI
1. Expert System
• In artificial intelligence, an expert system is a computer
system that emulates the decision-making ability of a
human expert.
• Expert systems is developed to mimic the expertise and
knowledge of an expert in a particular field.
• Expert systems are designed to solve complex problems by
reasoning about knowledge, like an expert through the
application of knowledge and rules.
• The focus is on Collection, creation and maintenance of
knowledge base
Expert System
• Examples of suitable domain areas:-
– Medical diagnosis (assist rather than replace Doctors)
– faults diagnosis (circuit boards in cars, aircrafts)
– Medical image analysis (screening mass x-rays for abnormalities)
– Prospecting for oil and minerals
– Finance (deciding whether to approve loans)
– Identification of plants, animals, agricultural pests/diseases and
chemical compounds.
– Spelling and grammar in documents applications.
NB/:
Expert systems are generally used to advise non-experts in situations
where a human expert in unavailable (for example it may be too
expensive to employ a human expert, or it might be difficult to reach
location).
Components of an Expert Systems.
An expert system is made up of three essential parts:
• A user interface - This is the system that allows a non-expert user to
query (question) the expert system, and to receive advice.
• An inference engine - This acts rather like a search engine, examining
the knowledge base for information that matches the user's query.
• A knowledge base - This is a collection of facts and rules. The
knowledge base is created from information provided by human
experts.
An empty knowledge base is referred to as an expert system shell
which is a set of programs which allow the building of an
expert system through the creation of knowledge and rules.
Decision trees
• Expert systems are programmed with a series of
logical rules to find a solution. Very basic expert
systems can use Boolean logic decision trees to come
to conclusions. Boolean logic has two possible
outcomes, true and false (or 1 and 0).
Inference rules
Inference rules (Rule base) - it is part of the Knowledge base. It has
all the logical rules known to the Expert System.
The inference rules are written as; IF…THEN statements which
describe rules in the knowledge domain by the process of chaining.
A simple example of inference rules might be:
IF X has gills THEN X is a fish
IF X is a fish THEN X lives in water
IF X lives in water THEN X can swim
Chaining:-The fundamental reasoning approaches implemented in
rule-based expert systems are Forward chaining (data driven) and
backward chaining (hypothesis driven) techniques.
Forward chaining starts with the data available and uses the
inference rules to extract more data until a desired goal is reached.
Backward chaining starts with a list of goals and works backwards to
see if there is data which will allow it to conclude any of these goals.
Read further:- Forward & Backward chaining tutorial
Expert System
Use of E.S in Diagnosing a disease
• Symptoms are entered by the user through the
interactive user interface /Asks questions...
• Expert system analyses data
• Inference engine compares symptoms with
those held in knowledge base using rules base.
• If matches are found
• Suggests probabilities/possible diagnoses.
Advantages
• Expert systems provide consistent answers.
• They never forget to answer a question.
• Reduces the time taken to solve a problem.
• A less skilled work force is needed, which gives the potential of saving
money,
• Allows areas of the world (e.g remote areas) access to expertise which
they could not normally afford.
• Expert advice available all the time
• Knowledge of expert staff can be captured to some extent before they
move on.
• Can be used as a training aid to increase the expertise of staff
• Makes rational decisions without any emotional overhead
• Does not get tired or overworked.
• Efficient way of getting answers as it does not involve additional help
staff e.g. automated help systems
• Natural language interface would make the expert system more human
friendly
Disadvantages
• They tend to lack common sense in some of the decision-making
processes. (a human user tends to notice obvious errors, whereas a
computer wouldn't)
• Data integrity & accuracy - Errors in the knowledge base or outdated
data can lead to incorrect decisions being made.
• It can be expensive to set up in the first place.
• Considerable training is necessary to ensure the system is used
correctly by the operators.
• Can't easily adapt to new circumstances (e.g. if they are presented with
totally unexpected data, they are unable to process it) hence needs
updating.
• Can be difficult to use (if the non-expert user makes mistakes when
using the system, the resulting advice could be very wrong)
NB: The reliability problem depends on the knowledge domain-fault
diagnosis suggesting an incorrect solution may cause inconvenience but
medical system suggesting incorrect treatment could cause a serious
health impact.
Problems with Expert Systems
• Watch the following Video
[Link]
Note down any new issue identified.
Examples of an expert systems
Task
The following sites have interactive expert system demos to help
you understand the types of knowledge domains suitable for
expert system creation. Look and work through as many of these
examples as you can.
• WebFlex Demos Online expert system examples including
insurance fraud detection and car ranking.
• Exsys Corvid Expert System Demos Online expert system
examples including a restaurant recommender and a dog breed
selector.
• Whale Watcher Identifies species of whale by their
characteristics.
• Work through this Car Fault Expert System Tutorial then try the
Demo at the end.
• NHS Direct Check Your Symptoms Online medical diagnosis
from the National Health Service.
• Cessna engine failure simulation - Citation Encore Expert
System)