QUESTION 1 a. Describe how expert systems perform inference [4 marks]

Inferencing is the reasoning process of AI. It takes place in the brain of an AI process.
The brain of an expert system is the inference engine that provides a methodology for reasoning about information in the knowledge base. Inference can be performed using semantics networks, production rules, and logic statements.

b. Describe generic categories of Expert Systems applications [16 marks]
Diagnosis. These are mainly used in the medical field to recommend remedies to illnesses, trouble-shoot electronic or mechanical problems or as debugging tools.

Repair. Expert systems that define repair strategies are also very common. As well as diagnosing the problem they can suggest a plan for the repair of the item. The repair plan typically contains a scheduling structure and some control structure to validate the repair process. Such systems have been employed in the automotive repair field and similar areas.

Instruction. They are mainly used in the training of pilots . Instructional expert systems have been used for individualised training or instruction in a particular field. The system presents material in an order determined by its evaluation of the user’s ability and current knowledge and monitor’s the progress of the student, altering the sequence depending on this progress.

Interpretation. Interpretive expert systems have the ability to analyse data to determine its significance or usefulness. The knowledge base often contains models of real world situations which it compares to its data. These are often used in exploration for mineral, gas and oil deposits as well as in surveillance, image analysis and speech understanding.

Predictive expert systems are used as a method to “guess” at the possible outcomes of observed situations, usually providing a probability factor. This is used often in weather forecasting.
Design and Planning. This allows experts to quickly develop solutions that save time. These systems do not replace experts but act as a tool by performing tasks such as costing, building design, material ordering and magazine design.

Monitoring and Control. In certain applications expert systems can be designed to monitor operations and control certain functions. These are particularly useful where speed of decision making is vitally important, for example in the nuclear energy industry, air traffic control and the stock market.

Classification/Identification. These systems help to classify the goals in the system by the identification of various features (these can by physical or non-physical) For example various types of animals are classified according to attributes such as habitat, feeding information, colour, breeding information, relative size etc. These systems can be used by bird watchers, fishing enthusiasts, animal rescue shelters (to match animals to prospective owners) to name a few.

QUESTION 2
Discuss the limitations of Expert Systems [30 marks]

-The major difficulty in developing these systems is extracting the expertise needed to develop the knowledge base. It is difficult to extract an experts’ knowledge and codify it into a format that can be used in an automated application. Sometimes experts they tend to withhold data or information so that they are in control of the system.
-Expert Systems differ from traditional DSS in that the knowledge base is more complex due to less reliance on the end-user to evaluate the results. The aspect of artificial intelligence, makes it even more complicated to the user.

-Artificial intelligence (expert Systems) are not creative (though lately this is open to debate), it is limited in the use of sensory devices (also subject to debate), it cannot make use of a very wide context of experiences, and it does not use common sense.

-Knowledge is not always readily available.

-It can be difficult to extract expertise from humans. Some expertise comes with experience, so to get that experience is not feasible. Some experts die with what knowledge they have acquired.
-There are frequently multiple correct assessments.
-Time pressures. Time may not always be there to be fully conversant with the system.
-Users have cognitive limits.
-Expert System works well only within a narrow domain of knowledge.
-Most experts do not have an independent means to validate results.
-Vocabulary is often limited and difficult to understand.
-Help from knowledge engineers is difficult to obtain and costly.
-Potential for lack of trust on the part of the end-users.
-Knowledge transfer is subject to biases.

-Technology can be used to facilitate decision- making, but managers must be part of the process and work with it. With the way technology is fast changing, managers should have literacy in technology in general and a bit more concerning expert systems to make sound and effective decisions.

It is good to capture expertise in a system so that if the "expert" leaves the knowledge does not go with him or her.

-There needs to be adaptability in any system that is built. Users should be able to adopt to the new system with ease and comfortably.
-No technology offers an easy and total solution. Large systems are costly and require significant development time and computer resources. -Maintaining human expertise in organizations sometimes is difficult because of the following reasons:-
• their knowledge is from a narrow field, don’t know the limits • the answers are not always correct (advices have to be analysed!) • don’t have common sense (greatest restriction) →all of the self-evident checking have to be defined

Sources:

http://www.scribd.com 2 July 2012 10:31am http://www.comsci.us/ai/notes/chap05.html 6 July 2012 12:00pm http://www.softcomputing.net/fuzzy_chapter.pdf 9 July 2012

Expert Systems 4207- PDF – Chamanga Notes 28 June 2012
7