Topic 181

access to the knowledge base underlying an inference engine in an expert system, for example, whether people affected by decisions made using an expert system should have access to the rules by which the decision was made. by Vaibhav

Access to the knowledge base underlying an inference engine in an expert system, for example, whether people affected by decisions made using an expert system should have access to the rules by which the decision was made.

To begin with let's recall what an expert system and inference engine are so an expert system is "a system that attempts to provide solutions to problems in a particular field, based on a database of information about that field. An expert system might specialize in biology, accounting, human resources, and financial service and be able to provide answers to questions on all these fields, and even carry out complex thought processes on questions relating to all these fields." And an inference engine is "the part of an expert system that draws inferences and deduces new facts by using old knowledge."

People affected by decisions made using an expert system should always have access to the rules by which the decision was made so that the process of attaining the decision can be justified when needed to. One reasonable way of looking at why there is a need for the rules is to look at the drawbacks of an expert system. To learn the drawbacks we must understand the practical use of an expert system first.

The lack of human common sense needed in some decision makings can sometimes provide unrealistic decisions. The creative responses human experts can respond to in unusual circumstances and can lead to various problems. Domain experts not always being able to explain their logic and reasoning The challenges of automating complex processes can lead to an undesired outcome. The lack of flexibility and ability to adapt to changing environments can cause problems in the field of biology. Not being able to recognize when no answer is available. For instance in biological cases the system may provide answers which are not possible and doctors would not instantly see that there is no real solution to the issue.

Expert system is used and applied in several medical cases. It is applied when the case is not brief and doesn't have a simple answer to it. Meaning it has a multiple solutions to it, therefore an expert system produces algorithmic approach. Hence a very narrow topic such as "diagnosing skin in human teenagers" can be used to provide rules on thumbs on how to evaluate the problem. The expert system will use an expert developer which would draw inferences from the old knowledge previously inputted by an expert in the field.

In conclusion, an expert system is always right but the rules on which it is run are inputted by an expert in the field who can be wrong from time to time. Hence if a wrong solution is provided especially in the field of biology and financial services someone who doesn't know much about the field or has absolutely no knowledge may choose to simply rely on the advice given by the system when it is wrong. In today's world where there are a number of ways to approach something, the best and expert in the field can do to is to make sure the validation, verification and evaluation of the system are made at their best.