Ontologies: Need, usage and attainment of health care system
This paper describes how we applied the PROTÉGÉ-II architecture to build a knowledgebased system that configures elevators. The elevator-configuration task was solved originally with a system that employed the propose-and-revise problem-solving method (VT; Marcus, Stout & McDermott, 1988). A variant of this task, here named the Sisyphus-2 problem, is used by the knowledge-acquisition community for comparative studies. PROTÉGÉ-II is a knowledge-engineering environment that focuses on the use of reusable ontologies and problem-solving methods to generate task-specific knowledge-acquisition tools and executable problem solvers. The main goal of this paper is to describe in detail how we used PROTÉGÉ-II to model the elevator-configuration task. This description provides a starting point for comparison with other frameworks that use abstract problem-solving methods. Starting from a detailed description of the elevator-configuration knowledge (Yost, 1992), we analyzed the domain knowledge and developed a general, reusable domain ontology. We selected, from PROTÉGÉ-II’s library of preexisting methods, a propose-and-revise method based on chronological backtracking. We then configured this method to solve the elevator-configuration task in a knowledge-based system named ELVIS. We entered domain-specific knowledge about elevator configuration into the knowledge base with the help of a task-specific knowledge-acquisition tool that was generated from the ontologies. After we constructed mapping relations to connect the domain and method ontologies, PROTÉGÉ-II generated the executable problem solver. We have found that the development of ELVIS has provided a valuable test case for evaluating PROTÉGÉ-II’s suite of system-building tools. 1 PROTÉGÉ-II AND SISYPHUS-2 To evaluate a general architecture for software development, developers must test the architecture with realworld tasks. The best way to validate the strengths and to discover the weaknesses of an architecture is to apply the ideas and tools of the architecture to a task that is similar in size and complexity to tasks found in the real world. The PROTÉGÉ-II architecture is a set of tools and a methodology for developing knowledgebased problem-solving systems. The Sisyphus-2 problem is a large-scale task of configuring elevator systems; it is a variant of the problem solved by the VT system (Marcus, Stout & McDermott, 1988). This paper describes how we used the PROTÉGÉ-II architecture to solve the Sisyphus-2 problem. The knowledge-acquisition research community selected the Sisyphus-2 task as a benchmark for comparing knowledge-modeling approaches, problem-solving methods, and reusability of knowledge structures. This elevator-configuration task consists of selecting appropriate components and dimensions to configure an elevator system according to user specifications and safety constraints. Our knowledge base for this task is primarily defined by the Sisyphus-2 document (Yost, 1992) and by a proposed ontology for configuration 2 Rothenfluh, Gennari, Eriksson, Puerta, Tu & Musen design and the Sisyphus-2 task.1 To provide an idea of the problem description, we list in Figure 1 the section headings of the Sisyphus-2 document (Yost, 1992). This document is pivotal to our work with Sisyphus-2 because it provided all the essential information, concrete data, and well-documented formulae in this complex domain. A secondary source of information about the elevator-configuration task, associated tools, systems, and problems is the literature about SALT, a knowledge-acquisition system developed for the original VT task (Marcus & McDermott, 1989). A core idea for the success of Sisyphus-2 is the use of ontologies that describe terminology and knowledge appropriate for the elevator-configuration task. Such ontologies should facilitate the comparison of problem-solving methods developed in different theoretical frameworks (for related issues about sharing and reuse, see the ARPA Knowledge-Sharing Effort—e.g., Neches, Fikes, Finin, Gruber, Patil, Senator, & Swartout, 1991). In PROTÉGÉ-II, we use the word ontology in the same sense as in Ontolingua (Gruber, 1993). PROTÉGÉ-II is a knowledge-engineering environment that enables developers to define knowledge-acquisition tools and knowledge systems by reusing problem-solving methods and domain ontologies. The PROTÉGÉ-II architecture emphasizes the automatic generation of knowledge-acquisition tools and performance systems from declarative, domain-oriented knowledge structures. The original description of SALT’s goals is almost identical to a description of our goals: SALT is a program that acquires knowledge from an expert and generates a domain-specific knowledge base compiled into rules. SALT then combines this compiled knowledge base with a problem-solving shell to create an expert system. SALT maintains a permanent, declarative store of the knowledge base which is updated during interviews with the domain expert and which is the input to the compiler/rule generator. It is this intermediate language which represents knowledge by function. (Marcus & McDermott, 1989, p. 3) 1 The text document as well as the ontologies and the knowledge bases are available as computer files through anonymous ftp from ksl.stanford.edu.