KNOWLEDGE ACQUISITION FROM MULTIPLE EXPERTS

11.5 KNOWLEDGE ACQUISITION FROM MULTIPLE EXPERTS

An important element in the development of an ES is the identification of experts. This is a complicated task, perhaps because practitioners use so many support mechanisms for certain tasks (e.g., questionnaires, informal and formal consultations, texts). These support mechanisms contribute to the high quality of professional output, but they may also make it difficult to identify a "knowledge czar" whose estimates, processes, or

P A R T I V INTELLIGENT DECISION SUPPORT SYSTEMS

TABLE 1

1.6 Benefits of and Problems with Participation of Multiple Experts

Benefits

Problems

On the average, fewer mistakes by a group of Groupthink phenomena experts than by a single expert Several experts in a group eliminate the need

Fear on the part of some domain experts of for using a world-class expert (who is

senior experts or a supervisor (lack of difficult to get and expensive)

confidentiality)

Wider domain than a single expert's Compromising solutions generated by a group with conflicting opinions

Synthesis of expertise Wasted time in group meetings Enhanced quality from synergy among

Difficulties in scheduling the experts experts

Dominating experts (controlling, not letting

others speak)

The usual approach to this problem is to build ES for a very narrow domain in which expertise is clearly defined; then it is easy to find o n e expert. H owever, even though many ES have b e e n constructed with one expert—an approach advocated as a good strategy for E S construction—there could be a need for multiple experts, espe- cially when more serious systems are being constructed or when expertise is not partic- ularly well defined. The case described in the Opening Vignette (Ranjan et al., 2002) is

a good example of using multiple experts. Table 11.6 lists the benefits and problems of multiple experts. The major purposes of using multiple experts are

To better understand the knowledge domain.

To improve knowledge-base validity, consistency, completeness, accuracy, and rel-

evancy.

To provide better productivity.

To identify incorrect results more easily.

To address broader domains. To be able to handle more complex problems and combine the strengths of differ- ent reasoning approaches.

W h e n multiple experts are used, there are often differences of opinion and con- flicts that must be resolved. This is especially true w h e n k n o w l e d g e bases are being

developed from multiple sources where these systems address problems that involve the use of subjective reasoning and heuristics.

Other related issues are identifying the various aspects of the problem and match- ing them to the appropriate experts, integrating k n o w l e d g e f r o m different experts,

assimilating conflicting strategies, personalizing c o m m u n i t y k n o w l e d g e bases, and developing programming technologies to support these issues.

MULTIPLE -EXPERT SCENARIOS There are four possible scenarios, or configurations, for using multiple experts (McGraw

and Harbison-Briggs, 1989; O'Leary, 1993; R a y h a m and Fairhurst, 1999; Scott et al., 1991): individual experts, primary and secondary experts, small groups, and panels.