Showing papers by "Riichiro Mizoguchi published in 1996"

A method of qualitative reasoning for model-based problem solving and its application to a nuclear plant

Abstract: Model-based expert systems are expected to contribute to overcoming the difficulties of conventional rule-based systems. This paper describes the modeling of mechanical systems and a method of qualitative reasoning based on causal specifications. The causal specifications represent a component's local causal properties, following the principles for reusability and composability. It contributes to providing intuitive causal ordering of complex behavior originated in the combination of components, including inter-component negative feedback. A model of a system is represented by combining a set of local component models and global knowledge derived from general properties of the physical entity. This allows for reusable knowledge which is easy to describe. Furthermore, the method has been successfully applied to a nuclear power plant. Reasoning results were unambiguous and matched those obtained by domain experts.

Identification of ontologies to reuse knowledge for substation fault recovery support system

Abstract: A methodology to identify task/domain ontologies is proposed in this paper According to the methodology, task/domain ontologies that are indispensable for building a reusable knowledge base are identified The expertise model of the methodology has two kinds of ontologies at different abstraction levels The first one is based on a vocabulary applicable in common to the entire target field This vocabulary is used as a communication basis between domain experts and knowledge engineers The first layer of an expertise model is described in this vocabulary and is built in collaboration with them The second one is based on a generic vocabulary By defining the common vocabulary in terms of the generic vocabulary, a reusable task model that is an expertise model at the second layer can be built We evaluate the effectiveness of this methodology by applying it into substation restoration problems

Modeling ill-structured optimization tasks through cases

Abstract: CABINS is a framework of modeling an optimization task in ill-structured domains. In such domains, neither systems nor human experts possess the exact model for guiding optimization, and the user's model of optimality is subjective and situation-dependent. CABINS optimizes a solution through iterative revision using case-based reasoning. In CABINS, task structure analysis was adopted for creating an initial model of the optimization task. Generic vocabularies found in the analysis were specialized into case feature descriptions for application problems. Extensive experimentation on job shop scheduling problems has shown that CABINS can operationalize and improve the model through the accumulation of cases.

A Qualititative Reasoning Based on an Ontology of Fluid Systems and Its Evaluation

Abstract: This research is concerned with causal understanding and qualitative reasoning of behavior of physical systems, which are crucial issues of modelbased problem solving. In this paper, a new method of qualitative reasoning and causal ordering is proposed and its application to a power plant is presented. The method is based on our kernel ontologies of causality and time-resolution and a domain ontology of fluid systems. These ontologies help make the design rationales of our method explicit and facilitate reusability of our models. The whole of the target system is represented by combining a set of local component models and global constraints. The component models include local and causal characteristics of each component which are independent of context for their reuse on the basis of the ontology of causality. Global constraints with time-scales are derived according to the general properties of the physical entity which are prepared beforehand as a part of the domain ontology. They contribute to providing intuitive causal ordering of complex behavior originated in various configurations of components, including inter-component negative feedback. Furthermore, the method has been successfully applied to a power plant. All the reasoning results matched those obtained by a domain expert including their ambiguities.

AI and ES research for DSS

Abstract: This special issue deals with the state-of-the-art collection of Artificial Intelligence and Expert Systems research for the Decision Support Systems. For this purpose, the nine best papers are selected from the proceedings of the 94 Japan/Korea Joint Conference on the Expert Systems which is now renamed as Pacific Asian Conference on Expert Systems (PACES). Since the contributors to the conference are not limited to the Far East region, the nationality of authors encompass not only Japan and Korea, but also Hong Kong, France, and the U.S.A. The first three papers deal with the issue of knowledge engineering which is the bottlenecked technology in developing expert systems. The first paper "Identification of Ontologies to Reuse Knowledge for Substation Fault Recovery Support System" (Y. Takaoka and R. Mizoguchi) proposes a methodology to identify the task/domain ontologies to help the reuse of knowledge base. The second paper "An Approach to Improving the Maintainability of Existing Rule Bases" (Kunihiko Higa) introduces the formal procedure to detect and simplify the complex rule structure. The third paper "Knowledge Acquisition as a Constructive Process: A Methodological Issue" (Bernard Le Roux) provides a principled approach to knowledge modelling and elicitation. Two papers utilize the neural networks. The paper "UNIK-OPT/NN: Neural Network Based Adaptive Optimal Controller on Optimization Models" (Wooju Kim and Jae Kyu Lee) provides an environment of building neural networks that can run on top of the optimization models for control purpose. The paper "Hybrid Neural Network Models for Bankruptcy Predictions" (Kun Chang Lee, Ingoo Hart, and Youngsig Kwon) proposes three hybrid neural network models for bankruptcy prediction. Two papers utilize the knowledge based technologies to the design of information systems. The paper "Merging CASE Tools with Knowledge-Based Technology for Automatic Software Design" (Behrouz H. Far, Mad Ohmori, Takeshi Baba, Yasukiyo Yamasaki, and Zenya Koono) develops the knowledge based software engineering tool by merging with a conventional CASE tool. The paper "An Expert System with Case-Based Reasoning for Database Schema Design" (Yong-Kee Pack, Jungyum Seo, and Gil-Chang Kim) proposes a relational conceptual graph to represent the users design specification and cases. The paper "IBRS: Intelligent Bank Reengineering System" (Daniel MoonKee Min, Jong Ryul Kim, Won Chul Kim, Dai Hwan Min, and Steve Ku) develops a knowledge based system which assists a bank in choosing the most appropriate business process engineering alternative, and the paper "INTERFACER: A User Interface Tool for Interactive Expert-Systems" (Shigeo Kaneda, Megumi Ishii, Fumio Hattori, and Tsukasa Kawaoka) resolves the user interface development bottleneck in building expert systems. We appreciate the authors' contribution in extending and revising the manuscripts.

An Inteligent Education System Which Supports Scientific Thinking: Galileo -Philosophy and Basic Architecture

Abstract: This paper discusses, Galileo, an Intelligent Education System which facilitates learners' scientific thinking. In the discussion we will focus on the concepts behind the designing of Galileo, and the guidance that it provides. First, we will discuss how to support scientific thinking. For this purpose, we will discuss naive knowledge which has been formed through learner's daily experiences. A learner who has naive knowledge cannot think about a given situation scientifically because the naive knowledge interferes with the appropriate understanding of the situations. Next, we present a design of Galileo focusing on teaching interaction and the learner models that enables it.

Discourse Management System for Communication Through Spoken Language

Abstract: Publisher Summary This chapter discusses the development of a speech understanding system called SPURT-I, which accepts utterances describing simple scenes In order to realize communication through spoken language, it has to accept discourse Speech understanding system is currently developing with a discourse management system One of the characteristics of discourse is that the user does not always respond to a question posed by the system When the system asks a question, for example, the user may ask another question instead of answering if he/she does not know what to answer or may answer incorrectly Furthermore, the representations of the answers take various forms Therefore, the system has to identify whether the utterance is an answer or a question and which question it corresponds to when it is an answer This chapter describes the current status of the development of a new version of the system The discourse management system takes care of the identification of the correspondence between a requirement and a response based on the stimulus and response (SR) plan and two kinds of stacks The interaction between the discourse management system and two other subsystems is also discussed in this chapter

Prediction of F0 parameter of contextualized utterances in dialogue

Abstract: In order to synthesize natural spoken dialogue, it is necessary to incorporate dialogue information into generation of the surface sentence and the prosody. The paper describes the prediction of the F0 maximum for minor phrases in dialogue based on a two-step predictive method. Special attention is directed to specific phrases containing the person's name or the day of the week in a schedule arrangement task in order to narrow the diversity of characteristics of F0 parameters in dialogue. Seven features were identified as dialogue information which are useful to predict the F0 parameter. Two D-rule sets derived from the person's name or the day of the week are very similar to one another. They reduce the total prediction errors by about 50% for the data which have much influence on dialogue context.