![]() |
|||
![]()
|
![]() |
![]() |
![]() |
1.1. EXPERT SYSTEMSExpert systems are systems that are capable of offering solutions to specific problems in a given domain or that are able to give advice, both in a way and at a level comparable to that of experts in the field. Building expert systems for specific application domains is known as knowledge engineering. The way that expert systems are built results in some unique advantages. Although expert systems are still expensive to build and maintain, they are inexpensive to operate. An expert system can be easily distributed in a number of copies, whereas training a new human expert is much more time-consuming and expensive. An expert system can reduce the information that human users need to process, reduce personnel costs, and increase throughput. Expert systems are likely to perform tasks more consistently than human experts. An expert system will handle similar situations in the same way and make comparable recommendations, whereas humans are influenced by various effects, such as recency and primacy effects. For example, most recent information has a disproportionate impact on one's judgment. On the other hand, primacy effects refer to the fact that early information dominates the judgment. An expert system can provide permanent documentation of the decision process. The knowledge of several human experts can be combined to give a system more breadth than a single person is likely to achieve. Expert systems can help a firm create entry barriers for potential competitors. A central task underlying many of the activities of attorneys is inferring the legal consequences of a given set of facts. Some systems use a case-based reasoning approach in which new cases are compared with the smallest collections of precedent facts that justified an individual inference step in the explanation of a precedent case. This sets a precedent to be used in case comparison. Case comparison also is assisted by an expressive semantic network representation of case facts. Techniques are presented for retrieving and comparing cases represented in this formalism. While expert systems have many advantages, they also have some weaknesses. For example, expert systems are not good at judgments that depend on meta-knowledge, i.e., knowledge about their own expertise. Expert systems are not good at recognizing when no answer exists or when the problem is outside their area of expertise. Human experts, in addition to technical knowledge, have common sense. Human experts automatically adapt to changing environments; expert systems must be explicitly updated. Human experts can respond creatively to unusual situations, while expert systems cannot. It is not yet known how to give expert systems common sense. 1.2. THE ACCEPTANCE AND SUCCESS OF AN EXPERT SYSTEMExpert systems (ESs) are used in an increasingly wider range of application areas. Companies and businesses more and more rely on expert systems. Such widespread use raises issues concerning their value and factors that may affect the success of incorporating an expert system in practice. Are existing evaluation techniques from the HCI applicable for assessing expert systems? It has been found that the success of ESs, in terms of user satisfaction, is directly related to the quality of developers and the ES shells used, end-user characteristics, and degree of user involvement in ES development. Several recommendations are proposed for ES project managers to enhance the likelihood of project success, including adding problem difficulty as a criterion for ES application selection; increasing ES developer training to improve people skills; having the ability to model and use a systems approach in solving business problems; sharpening end-user attitudes and expectations regarding ES; improving the selection of domain experts; more thoroughly understanding the ES impact on end-user jobs; restricting the acquisition of ES shells based on a set of criteria; and ensuring a proper match of ES development techniques and tools to the business problem at hand. 1.3. THE ROLE OF ITS USER INTERFACEUser interfaces need to deal with three fundamental problems: communication, control, and access. The first problem is about communication between users and the system. The second problem is about control and the issue of who does what when and task allocation in general. The third problem concerns making the full range of facilities and computational capabilities accessible and useful to users. 2. BACKGROUNDThe acceptance of an expert system by the end-user has been regarded as one of the major criteria for the success of an expert system. An expert system needs to be easy to learn and friendly to use. After all, it must fit into the existing work environment and be compatible with other computer systems in use. It is also known that user interface design is one of the main reasons that hindered expert systems transition from prototype into everyday use. The success of expert systems therefore depends upon the user interfaces as well as the efficiency of its knowledge encoding and reasoning. 2.1. HISTORYThere are an ever-growing number of books and papers on expert systems. Several classic ones are discussed in Clancy and Shortliffe (1984). The first expert systems were developed as early as the late 1960s. The research in expert systems gained substantial growth in the 1970s. The early expert systems mostly concerned the field of medical diagnosis. The best-known expert system in medicine is MYCIN, developed in the 1970s at Standford University. The MYCIN system is able to assist internists in the diagnosis and treatment of a number of infectious diseases. This system has given an important impulse to the development of similar expert systems in fields other than medicine. Another classic example is XCON, previously called R1, which can configure computer systems from Digital Equipment Corporation (DEC) such as VAX and PDP11. Configuring a computer system requires considerable skill and effort. In the late 1970s, DEC and Carnegie-Mellon University started the development of XCON. XCON has been fully operational since 1981.
|
![]() |
|
Use of this site is subject certain Terms & Conditions. Copyright (c) 1996-1999 EarthWeb, Inc.. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of EarthWeb is prohibited. Please read our privacy policy for details. |