详情

原文传递 MVP-CA METHODOLOGY FOR THE EXPERT SYSTEM ADVOCATE'S ADVISOR (ESAA).

题名：	MVP-CA METHODOLOGY FOR THE EXPERT SYSTEM ADVOCATE'S ADVISOR (ESAA).
作者：	Mehrotra-M
关键词：	KNOWLEDGE-BASED-SYSTEMS; SOFTWARE-; CLUSTERING-; VALIDATION-; EXPERT-SYSTEMS; SYSTEM-ARCHITECTURE; REDUNDANCY-; INCONSISTENCIES-; ANOMALIES-; IDENTIFICATION-
摘要：	The Multi-ViewPoint Clustering Analysis (MVP-CA) tool is a semi-automated tool to provide a valuable aid for comprehension, verification, validation, maintenance, integration, and evolution of complex knowledge-based software systems. In this report, the contractor has shown the value of using their MVP-CA tool for analyzing a poorly structured rule base such as the Expert System Advocate's Advisor (ESAA). In particular, they have shown that MVP-CA methodology is capable of exposing the current underlying software architecture of the knowledge base. Such capability can pave the way towards showing alternate ways of restructuring the system while the system is evolving. Also, the MVP-CA technology juxtaposes rules with similar content and structure so that incomplete specifications on various variables and qualifier values in the rules surface quite easily. Through grouping of rules from multiple perspectives, this report also shows how redundancies, inconsistencies, and anomalies become apparent easily. MVP-CA methodology stresses the proper structuring of a complex system first, so that understanding the system becomes a basis for verification, validation, testing, and maintenance.
总页数：	1997/11. 9412-9503 pp89 (2 Fig., 3 Tab., 7 Ref., 1 App.)
报告类型：	科技报告