حل تعارض در یک سیستم مبتنی بر دانش با استفاده از ویژگی های متعدد تصمیم گیری
|کد مقاله||سال انتشار||تعداد صفحات مقاله انگلیسی||ترجمه فارسی|
|18839||2009||7 صفحه PDF||سفارش دهید|
نسخه انگلیسی مقاله همین الان قابل دانلود است.
هزینه ترجمه مقاله بر اساس تعداد کلمات مقاله انگلیسی محاسبه می شود.
این مقاله تقریباً شامل 6130 کلمه می باشد.
هزینه ترجمه مقاله توسط مترجمان با تجربه، طبق جدول زیر محاسبه می شود:
- تولید محتوا با مقالات ISI برای سایت یا وبلاگ شما
- تولید محتوا با مقالات ISI برای کتاب شما
- تولید محتوا با مقالات ISI برای نشریه یا رسانه شما
پیشنهاد می کنیم کیفیت محتوای سایت خود را با استفاده از منابع علمی، افزایش دهید.
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 36, Issue 9, November 2009, Pages 11552–11558
The metarules useful in the conflict resolution are often directly related to the multiple, conflicting, and non-commensurate objectives associated with the problem domain. However, in its current form, the use of metarules for conflict resolution has some drawbacks. Above all, the metarule use in rule selection is not tailored to the current situation or the specific user; it is tailored to the domain expert, whose domain expertise and preferences were used to construct a knowledge-based system. In this paper, we presents a new method for resolving the conflicts of rules in the knowledge-based system, using decision analysis techniques that explicitly incorporate a user’s preference judgments about the rules. To this end, we consider a conflict resolution problem as a multiple attribute decision-making problem. Further, the proposed method allows for user’s preference judgments that are specified not by rigid format but by user-friendly format for the purpose of reducing burden of information specification. We have applied the proposed methodology to an organizational information-oriented service and resource planning, in which there exist multiple conflicting objectives to be considered.
Techniques from the disciplines of artificial intelligence (AI) and decision analysis (DA) have both been used extensively in the development of computerized decision aids. The expert system (ES), one of several methods for emulating human decision-making, is a computer program for solving decision problems that requires significant human expertise by using explicitly represented domain knowledge and computational decision procedures (Kastner & Hong, 1984). In normative DA, decision models have been developed to represent complicated real-world decision problems. Both ES and DA, though rooted in different fields, have basic conceptual similarities in terms of objectives (decision-aiding), delivery vehicle (the computer), and conceptual basis (graphs and networks). Unfortunately, from the perspective of many types of practical decision-aiding applications, both normative decision aids and expert system technology have significant limitations. Particularly, in the expert system development, there is a lack of established techniques for problem structuring and knowledge engineering. This usually leads to time consuming rule-based development efforts with limited success in domains where the knowledge required to solve problems is not already well established (Davis, 1982). Further, current expert systems do not explicitly consider preferences, which play a key role in DA; when preferences are (indirectly) addressed, they are the preferences of the expert, rather than the user’s preferences adjusted to the current problem solving environment (White, 1990). Normative decision analysis, on the other hand, is usually built around a prescriptive and rigid problem structure called a decision analysis model. This model, in turn, may not be compatible with the evolutionary approach to system development which is characteristic of AI (Lehner, Mattew, & Donnell, 1985). In this paper, we focus on the DA’s prescriptive methodology (prioritizing or ranking alternatives or options in a prescriptive manner), which incorporates a user’s preference judgments, for conflict resolution in a knowledge-based system. The process of using a knowledge-based system consists of a cycle having three phases: matching, conflict resolution, and action. Using inference, it is possible to identify a set of rules that matches the context. If this occurs during the matching process in a production system, some approaches should be applied to resolve possible conflicts (Barr and Feigenbaum, 1981, Davis and King, 1977 and Hayes-Roth et al., 1983). Metarules useful in conflict resolution are often directly related to the multiple, conflicting, and non-commensurate objectives associated with the problem domain. However, in its current form, the use of metarules for conflict resolution has three key drawbacks: 1. Use of metarules in rule selection is not tailored to the current situation or the specific user; it is tailored to the domain expert, whose domain expertise, preferences, etc. were used to construct the system. 2. Use of metarules in rule selection does not take into account objectives tradeoffs. 3. Metarules do not permit efficient representations of expert knowledge when many objectives or decision contexts must be taken into account. We propose DA-based methodology that integrates the two disciplines synergistically and further considers a group of individuals in a situation where multiple experts take part in. Finally, we have applied the proposed method to a real-world problem, information-oriented service and resource planning of a company where there exist multiple conflicting objectives under consideration. The rest of this paper is organized as follows. Section 2 reviews a number of approaches towards decision theoretic expert systems. Section 3 suggests the use of decision analysis techniques for conflict resolution in expert systems, and application are discussed in Section 4.