It is interesting to note that there is no agreed upon definition of artificial intelligence. Why is this interesting? Because government agencies ask for it, software shops claim to provide it, popular magazines and newspapers publish articles about it, dreamers base their fantasies on it, and pragmatists criticize and denounce it. Such a state of affairs has persisted since Newell, Simon and Shaw wrote their first chess program and proclaimed that in a few years, a computer would be the world champion. Not knowing exactly what we are talking about or expecting is typical of a new field; for example, witness the chaos that centered around program verification of security related aspects of systems a few years ago. The details are too grim to recount in mixed company. However, artificial intelligence has been around for 30 years, so one might wonder why our wheels are still spinning. Below, an attempt is made to answer this question and show why, in a serious sense, artificial intelligence can never demonstrate an outright success within its own discipline. In addition, we will see why the old bromide that "as soon as we understand how to solve a problem, it's no longer artificial intelligence" is necessarily true.

The Artificial Intelligence in Medicine (AIM) Workshop has become a tradition. Meeting every year for the past nine years, it has been the forum where all the issues from basic research through applications to implementations have been discussed; it has also become a community building activity, bringing together researchers, medical practitioners, and government and industry sponsors of AIM activities. The AIM Workshop held at Fawcett Center for Tomorrow at Ohio State University, June 30 - July 3, 1984, was no exception. It brought together more than 100 active participants in AIM.

The Air Force Institute of Technology (AFIT) provides master's degree education to Air Force and Army Officers in various engineering fields It is in a unique position to educate and perform research in the area of applications of artificial intelligence to military problems. Its two AI faculty members are the only military officers with PhD's in Artificial Intelligence. In the past two years, the artificial intelligence Laboratory of AFIT has become a major focal point for AI research and applications within the government. In this article, we describe our on-going applications research in the areas of automated cockpit systems, natural language understanding, maintenance expert systems, expert systems for planning and knowledge based software design.

About a year ago. I was approached by Phoebe Huang of Comtex Scientific Corporation who hoped that I would help devise a dramatically expanded index for topics in AI to aid Comtex in indexing the series of AI memos and reports that they had been gathering. Comtex had tried to get the ACM to expand and update its classification. But was told that ACM had just revised the listing two years ago or so ago, and did not intend to revise it again for a while: even if they did. The revision might require a year or more to complete. Comtex wanted the new classification within six to eight weeks. I agreed to take on the task, thinking it wouldn't be too hard. The major decision I had to make was whether to use the existing ACM index scheme and add to it, or start with a fresh sheet of paper and devise my own. I decided to stick with ACM's top two levels, only adding, not modifying, major headings.

The Air Force Institute of Technology (AFIT) provides master's degree education to Air Force and Army Officers in various engineering fields It is in a unique position to educate and perform research in the area of applications of artificial intelligence to military problems. Its two AI faculty members are the only military officers with PhD's in Artificial Intelligence. In the past two years, the artificial intelligence Laboratory of AFIT has become a major focal point for AI research and applications within the government. In this article, we describe our on-going applications research in the areas of automated cockpit systems, natural language understanding, maintenance expert systems, expert systems for planning and knowledge based software design.

What are the powerful new ideas in knowledge based design? What important research issues require further investigation? Perhaps the key research problem in AI-based design for the 1980's is to develop better models of the design process. A comprehensive model of design should address the following aspects of the design process:the state of the design ; the goal structure of the design process;design decisions; rationales for design decisions; control of the design process; and the role of learning in design. This article presents some of the most important ideas emerging from current AI research on design especially ideas for better models design. It is organized into sections dealing with each of the aspects of design listed above.

Tin Nguyen performed the work contained in the article "Knowledge Base Verification" while at Lockheed and is currently working for Bell Northern Research as a member of the research staff. Deanne Pecora, a staff engineer with the Lockheed Artificial Intelligence Center, 2710 Sand Hill Road, Menlo Park, California 94025, is working on Rick Brigs, author of "Knowledge Representation and Inference in Sanskrit: A applying knowledge-based systems to Review of the First National Conference," is a senior engineer at Delfin Systems, real problems. She is a coauthor of 1349 Moffett Park Drive, Sunnyvale, California 94089. Briggs is currently working'Knowledge Base Verification." Walt Perkins, coauthor of "Knowledge Base Verification" is a consulting scientist Lindley Darden, who wrote "Viewing the History of Science as Compiled Hindsight," with the Lockheed Artificial is an associate professor in the departments of philosophy and history and Intelligence Center, 2710 Sand Hill a member of the graduate faculty in the Committee on the History and Philosophy Road, Menlo Park, California 94025 of Science at the University of Maryland, College Park. She is currently and the principal developer of the serving in the second year of a halftime research appointment at the University Lockheed expert system. of Maryland Institute for Advanced Computer Studies. Her mailing address is Department of Philosophy, University of Maryland, College Park, Maryland David Prerau is a principal member of 20742. The primary responsibility is to lead the author of "The 1985 Workshop on Distributed Artificial Intelligence, he is currently development of major expert systems working in the area of distributed artificial intelligence and is organizing with high corporate payoff and impact.

A pattern recognition algorithm is described that learns a transition net grammar from positive examples. Two sets of examples -- one in English and one in Chinese -- are presented. It is hoped that language learning will reduce the knowledge acquisition effort for expert systems and make the natural language interface to database systems more transportable. The algorithm presented makes a step in that direction by providing a robust parser and reducing special interaction for introduction of new words and terms.

In his recent article in AI Magazine, "AI prepares for 2001," Nils Nilsson put forward a paradigm of AI based Sufficiency implies finding a guide to investigate the on a declarative representation of knowledge with semantic case of human beings. I would like improve problem-solving performances succeeds only because to present some ideas and concepts stemming from current syntax mirrors semantics in the domains where the research in Genetic Epistemology (GE), initiated by Jean programs were applied. Piaget, there is then no need for any distinction between This interrogation is precisely the core of the Piagetian rules and metarules or knowledge-base and inference engines. The "epistemic program" should undergo by itself a GE is concerned with knowledge considered as a process, series of revisions of represeutations, and thus experiment [Piaget (1964)]. The obvious point of convergence different schemes of perceptions-or inference enginesas of AI and GE is precisely this concept of knowledge as a the mathematico-logical structure underlying the dynamic process.

The certainty-factor (CF) model is a commonly used method for managing uncertainty in rule-based systems. We review the history and mechanics of the CF model, and delineate precisely its theoretical and practical limitations. In addition, we examine the belief network, a representation that is similar to the CF model but that is grounded firmly in probability theory. We show that the belief-network representation overcomes many of the limitations of the CF model, and provides a promising approach to the practical construction of expert systems.