Showing papers in "Knowledge Based Systems in 1991"

Argument-based explanation of logic programs

Abstract: The paper argues that a satisfactory explanation of a logic program must take the form of an argument, rather than a proof. This can only be done on the basis of information regarding the role of the various literals in the bodies of the clauses, which is normally not captured by such programs. A schema for arguments, derived from Toulmin, is presented, and the components of this schema are related to the roles of literals in the bodies of clauses. A metainterpreter is described that uses annotations of body literals to build up an argument structure according to this schema. This structure can then be used to present the argument in a variety of ways; this is illustrated by a discussion of how the argument structure can be used as the basis of a presentation as a paragraph of text. A simple example from a quasilegal domain is presented.

Assembly sequence planning using case-based reasoning techniques

Abstract: Future computer aided design (CAD) systems will be not only smart, but also efficient in terms of solution strategies as well as time costs. This paper presents a problem-solving paradigm, namely case-based reasoning (CBR), and shows how it solves some constraint-oriented design problems efficiently. CBR solves a new problem by retrieving from its case library a solution which has solved a similar problem in the past and then adapting the solution to the new problem. The efficiency of such a system relies on the completeness and compactness (small size) of such a case library. These two characteristics of a case library in turn rely on an indexing scheme for the set of cases in the library. How these issues can be solved and how a case-based reasoning system can be designed for such problems is illustrated here, and the authors' theory is presented as it is applied to the area of assembly sequence planning.

Integrating expert systems and decision-support systems: principles and practice

Abstract: Commercial applications of artificial-intelligence technology often entail integration between information systems already in use and AI programs that are usually expert systems. Substantial problems arise in the different phases of the integration process. An expert system was developed that performed reasoning in the field of financial decision making, and it was embedded in a decision-support system that was already in use in a multinational manufacturing company. The first part of the paper presents a schematic representation of some problems with decision-support systems, and discussion of how knowledge-based systems could be used to solve them. The second part of the paper describes the methodological approach, and actual experience in the development of the project.

Knowledge-based system for material selection for design with new materials

Abstract: Designers expect more than support for the generation of drawings from the future generation of CAD systems, i.e. they need to be assisted by computer-aided decision making or applications working with artificial intelligence (AI). AI methods are particularly suitable for the processing of unstructured scattered knowledge for the solution of complex problems. One such problem is the selection of materials for construction with new materials (e.g. fibre-reinforced composite materials). The number of possible combinations and configurations of these materials is too large for the user, and, in particular, a nonexpert user, to be able to find optimal solutions. In the course of various research programs, a prototype of a knowledge-based system for material selection for design with new materials has been developed. Data about the materials and their properties are processed and stored in a database system. A computer-aided model has been developed that enables material data to be accessed via a comfortable graphical user interface. Selection of the optimal materials is possible through experience and expert knowledge (for example about production technology for fibre-reinforced composite materials) recorded as rules and facts in a database system. The expert-system shell Nexpert Object provides an appropriate mechanism to store and process expert knowledge. The developed expert system supports the designer in selecting the optimal material, observing given technical standards of the cured component. It selects alternatives, taking technical and economic issues into account, evaluates them, and offers them to the designer, who makes the final decisions.

Rigorous approach to the development of knowledge-based systems

Abstract: Knowledge-based systems have often been criticized for the limited theoretical base on which they are constructed The partially valid view, held by many, is that systems are often constructed in an ad hoc, individual way, which leads to unmaintainable, unreliable, and unrigorous systems This holds even though there have been several attempts at producing development methodologies to assist the knowledge engineer in the construction process1-7 A large contributing reason for the limited applicability of these methodologies is that they often too closely follow the waterfall model approach used for the development of conventional software systems8 This approach forces developers to make large jumps in the system state during development, which is not necessarily the most conductive way to model the domain accurately The paper therefore aims to introduce a new alternative approach that shows the benefits of taking a software engineering philosophy towards the development of knowledge-based systems The methodology breaks down the process of creating a knowledge-based system into constituent parts and discusses ways of creating rigorous specifications for those parts, as applicable This includes specifying the knowledge base, the representation, the control architecture, etc, thus promoting quality systems that are better specified, more reliable, and easier to maintain

Limited-resource scheduling by generalized rule-based system

Abstract: A general paradigm for solving resource allocation, time-tabling, and scheduling problems is presented. The paradigm is based on an expert system approach, which looks for a feasible solution that satisfies the problem's real-life constraints. The new paradigm includes generic concepts for resources, activities, constraints, and allocations. The general control strategy of the new paradigm is suitable for a large family of resource allocation and time-tabling problems. This control strategy includes three parts that deal with allocation, constraint checking, and changes to allocations. The proposed paradigm was tested on three real-world problems: crew assignment to air force missions, class scheduling for a university department, and time-tabling of final examinations for the faculty of natural sciences. All cases were solved well in a few minutes of central processing unit time, by Prolog-based systems that implemented the proposed paradigm. These case studies are described in the paper in some detail, and an overall evaluation of the methodology is given.

Knowledge-based programming environments

Abstract: This is a survey presented at the International Joint Conference on AI (IJCAI-89) in Detroit on Soviet work in the field of knowledge-based programming. It considers work which started in the 1970s as automatic programming projects with a goal to find new ways of program construction and, particularly, the work on computational models and problem solving. The structural program synthesis and conceptual programming technique developed in this framework were later used in the new generation computer project, START.

Towards a fuzzy logic programming system: a fuzzy propositional logic

Abstract: This paper formally develops a type of propositional fuzzy logic which is analogous to traditional two-valued logic. A complete set of fuzzy qualifiers are defined for the first time, and as a consequence a corresponding set of modifiers. In addition, a set of rules for combining those qualifiers in a proposition have been developed. Once these rules have been established, a definition of formulae in the context of propositional fuzzy logic is proposed and a theory of inference is deduced.

Logic-based approach to expert systems in chemistry

Abstract: A logic-based approach to expert systems for chemistry is discussed. In chemistry, most of the knowledge is expressed with chemical structural formulas. However, it is difficult to handle chemical structural formulas by using currently available logic programming languages. The logic programming language has been extended so as to include chemical structural formulas. This extended language/system named Chaus is presented in this paper. With Chaus language, the chemical knowledge is expressed in a natural way. The implementation to ensure practical efficiency and an application for drug design is also discussed.

Application of an expert system to monitoring and control in aquaculture

Abstract: When any measurement task is approached from the point of view of the knowledge which is needed, it is found that this knowledge often comprises a range of data, interpreted in terms of a set of rules. In the current explanatory application, measurement and control of water quality is fundamental to the breeding of healthy fish in aquaculture installations. It is the rule base, fed with real-time measured data, which is of primary importance in ensuring the health of the fish. Computer based expert systems simplify the application of rules and heuristics to real-time monitoring and control. This paper discusses the development of a computer-based, distributed monitoring and control system built around the "Crystal" expert system. The system enables both heuristic and real-time varying knowledge to be integrated into the rule base on which control decisions are automatically made. An important feature is that, by directly accessing the expert system's rule builder, the application domain user can change the configuration and operating rules of the system using only low-level computing skills. By also building 'sensor knowledge' into the expert system, the sensor design and operating requirements are simplified, allowing the inexpert user to specify the needed sensors, or build them to drawings supplied by the system.

Rule-based algorithms for geographic constraints in a marine knowledge-based system

Abstract: A suite of ruled-based algorithms designed to facilitate intelligent interaction with a computer representation of a nautical chart is described. The algorithms were developed for use in a prototype navigational knowledge-based system developed by a UK Science and Engineering Research Council funded research team at Liverpool Polytechnic, UK. The system was designed to provide decision support in the domain of marine navigation by providing the 'on-watch' navigator with advice on how best to avoid marine collisions with other vessels and land masses. An important attribute of the system is its ability to take into consideration geographical constraints when it is generating the advice. This is facilitated by the algorithms described. These are considered to be of general interest, as they represent a knowledge-based approach to well tried geographic-information system techniques that have to date been manipulated by the use of nondeclarative methods.

Some knowledge-acquisition methods for prospector-like systems

Abstract: The paper deals with the knowledge-acquisition methods designed and tested under the FEL-EXPERT Project, which is aimed at the development of rule-based diagnostic shells. Three different approaches have been used for knowledge acquisition: pattern-recognition, decision-tree, and intensional, pure probabilistic approaches. The designed methods may be applied to a wide subclass of rule-based diagnostic systems that exploit the pseudo-Bayesian model for uncertainty handling. The experimental results are discussed.

Automatic refinement of knowledge bases with fuzzy rules

Abstract: The knowledge-acquisition bottleneck obstructs the development of expert systems. Refinement of existing knowledge bases is a subproblem of the knowledge-acquisition problem. The paper presents a HEuristic REfinement System (HERES), which refines rules with mixed fuzzy and nonfuzzy concepts represented in a variant of the rule representation language Z-II automatically. HERES employs heuristics and analytical methods to guide its generation of plausible refinements. The functionality and effectiveness of HERES are verified through various case studies. It has been verified that HERES can successfully refine knowledge bases. The refinement methods can handle imprecise and uncertain examples and generate approximate rules. In this aspect, they are better than other famous learning algorithms such as ID3 15–18 , AQ11, and INDUCE 14, 19, 20 because HERES' methods are currently unique in processing inexact examples and creating approximate rules.

DEXT: an integrated knowledge-engineering environment for control-engineering applications

Abstract: The paper presents dext , an integrated knowledge-engineering tool for control-engineering applications. First, the domain of control engineering has been analysed so that a few of the key requirements to be satisfied by the tool may be identified. A large number of software packages developed over the years for dealing with different aspects of control engineering constitute the primary knowledge base of an expert system for such applications. Unfortunately, these packages have been developed in different languages, and they differ in their performance with regard to computational efficiency and numerical stability. Hence, transparent access to a bank of such packages must be allowed, and such accesses must be independent of the language in which the packages are written. Apart from providing such an interface, the proposed tool must also support a variety of data types. A structured organization of the design information is also required, as the expert system may refer to such facts at different levels of abstraction. The structure of the rules and the facts is described in detail. The salient features of dext include the feature of transparent interfacing with the external packages, and enhancement of inferencing efficiency through the techniques of rule partitioning and intelligent backtracking.

Representing temporal knowledge in conceptual graphs

Abstract: This study was motivated by some difficulties encountered by the authors when trying to express temporal knowledge using Sowa's conceptual graph (CG) approach. An overview of Sowa's approach is given and the difficulties encountered when trying to model temporal knowledge are outlined: the disparity of notations allowed by CG theory for expressing temporal information; the ambiguity and incompleteness of tense sspecification; the difficulty of harmonizing tenses and intergraph temporal relations. Various approaches suggested for representing time both in artificial intelligence and linguistics are presented, and an extension to Sowa's approach is proposed in which temporal and nontemporal knowledge are differentiated. In this model points in time are represented as well as time intervals. A semantic interpretation of verbs is provided based on an extension of Reichenbach's model of temporal markers. The authors show how their approach enables the representation of tenses as well as the aspectual properties of natural language sentences.

Model-theoretic approach to clustering

Abstract: The paper deals with a model-theoretic approach to clustering. The approach can be used to generate cluster description based on knowledge alone. Such a process of generating descriptions would be extremely useful in clustering partially specified objects. A natural byproduct of the proposed approach is that missing values of attributes of an object can be estimated with ease in a meaningful fashion. An important feature of the approach is that noisy objects can be detected effectively, leading to the formation of natural groups. The proposed algorithm is applied to a library database consisting of a collection of books.

Rule by example: an example-based approach to knowledge-base management

Abstract: An example-based approach to knowledge-base management involves capturing all rules as example data. Such an approach has many advantages, as data have been studied extensively, and a variety of techniques for effective management have been developed. Expressing all rules as stereotypical data allowed in the system provides a comprehensive solution to the three basic problems of a knowledge-based system. It provides a simple end-user language by reducing rules to data, which have simpler semantics; it allows the storage of rules as data, and hence the use of database technology to store, retrieve and maintain rules; and it simplifies the execution of rules by reducing the execution to data comparison.

Frame-based implementation of a design knowledge-capture scheme

Abstract: A knowledge-based system approach to a design knowledge capture (DKC) problem is presented. Such a system would capture not only the traditional definitive knowledge which describes a design, but also the explanatory knowledge which explains how the design decisions were made. Typical design processes were examined to determine what knowledge to retain and at what stage that knowledge is used. An outline of a DKC system was implemented using the Forms frame system. Although the problem of determining an optimum knowledge decomposition scheme is complex, a particular knowledge decomposition scheme useful to the design of magnetic bearing-based systems was imposed. It was determined that a DKC system is an extremely valuable engineering tool for complicated systems and systems with long service life.

ExpertPriz — an integrated expert system shell

Abstract: In this paper, hybrid expert system building tools are discussed. The PC-based expert system shell Expert-Priz, allowing shallow and deep knowledge to be coupled in a quite simple way, is introduced. ExpertPriz includes an inductive expert subsystem and a computational (conceptual programming) subsystem (solver) to handle shallow (expert) knowledge and deep (conceptual) knowledge, respectively. These subsystems can be used separately as well as in cooperation. The solver and expert subsystem and corresponding knowledge representation methods are described in the paper. The two interaction modes between these subsystems, ‘expert needs solver’ and ‘solver needs expert’, are discussed and illustrated by an example.

C-Priz — a new generation software tool for small computers

Abstract: This paper presents the C-Priz knowledge-based programming environment Computational problems can be solved automatically by C-Priz using program synthesis on computational models that can be built from concepts — pieces of knowledge The C-Priz language for specifying computational models (concepts) is described Dialogue scenarios built on top of the models merge two kinds of knowledge: concepts and expert knowledge about their usability Scenarios provide also a user-friendly graphical interface for applications The C-Priz environment toolkit for editing concepts, scenarios and graphics is described A sample C-Priz application is discussed

Analysis for knowledge-based systems: a practical guide to the KADS methodology: F R Hickman, J L Killin, L Land, T Mulhall, D Porter and R M Taylor Ellis Horwood Ltd, UK (1989) £24.95, ISBN 0 7458 0689 9, 186pp

Abstract: This book describes KADS, a structured methodology for the design of knowledge-based systems developed under the CEC ESPRIT collaborative research programme. It consists of a much researched approach to knowledge analysis and KBS design. KADS sees KBS design as a process of transformations of models of information that should be more structured than many iterative prototype-and-test approaches. The book starts, therefore, with a justification for a structured KBS methodology and moves into a general description of the area of KBS development techniques and tools. KADS attempts to avoid the technology-driven problem of much of knowledge engineering where the knowledge that is elicited is severely constrained by the representations the KBS development tool can accept. It does this by providing a library of abstract, generic knowledge structures. For a given task type the appropriate structure can be chosen and it is this that defines the knowledge that must be elicited and how it is to be represented. The book focuses on the tasks involved in this process and describes how the library is used. Two case studies are included that illustrate how the library of structures can be used and how one can generate an addition to the library when a suitable one is not available. One of the main aims of the book seems to be to attempt to convince knowledge engineers of the need for a structured method to KBS development. The authors present a clear argument for why it is not sufficient just to hack up a prototype and then to iteratively improve on this. They see prototyping as a useful technique within a well-managed methodology rather than a methodology in its own right. The book emphasizes what KADS calls the 'internal' requirements that relate to the knowledge base and provides little information about how the requirements of the users are dealt with ( 'external ' requirements) and how choices are made between different models of user-KBS interaction ( 'modality'). This suggests that KADS, itself, does not address this area as well as it might. As the importance of these areas in the take-up of knowledgebased systems is increasingly being recognized, it is this reviewer's opinion that this area needs to be given more attention in books like this. The book claims to be aimed at practising knowledge engineers, KBS vendors and students and teachers of KBS methods and to provide a practical guide to the techniques of KADS. While the level of detail is ideal for someone wanting an introduction to the KADS methodology and KBS development methods and issues in general, it cannot really be said to provide a working knowledge of KADS. As such, there is not really an acknowledgement of the difficulties that must be overcome in using the KADS approach, or any KBS development approach; indeed, it could suggest to the lay reader that such a methodology removes most of the problems. This is not to say that any pretence is intended. Rather, it is a question of the detail of analysis that is provided. For a book of its size it does its best to convey the areas in which the difficulties still lie. I found the book very readable and the standard of presentation is high. As an introductory text to KBS development it is on the expensive side, but it does make a valuable contribution to the literature, making it probably worth obtaining.

Reappraisal of the use of conditional probability in early expert systems

Abstract: A qualitative method of reasoning under uncertainty using Bayesian probability was originally proposed by Duda et al. However, this has been found to be 80% reliable at best in practice, and better techniques are being looked for by those who seek greater accuracy. The paper concludes that Bayesian approaches should not be abandoned. A Bayesian approach can be used as a sound quantitative technique, provided that the axioms of probability are properly observed and the nature of Bayes' theorem itself is understood. The paper documents some lessons that can be learned from the previous use of Bayes' theorem, and assists in the understanding of the way that it works in practice. The limitations of assuming conditional independence are emphasized, and the consequent equal ratios that must exist between various pairs of joint probabilities are highlighted. This leads to validity checks that can be made by an expert-system designer using local calculations and Bayesian update methods. It is also demonstrated that, for even modest numbers of antecedents, the assumption that these are independent overwhelms the information given by the expert, who, when considering a particular piece of evidence, has control of little more than the 'all other antecedents true' and the 'all other antecedents false' scenarios. The same conclusions may be pertinent to more recent work, but this is not explored in depth.

Knowledge-based approach to Petri nets analysis

Abstract: In this paper a knowledge-based experimental tool for Petri nets modelling and analysis is presented. The tool consists of a graphical editor, allowing the capture and editing of Petri net graphs, integrated with an expert system allowing an automated analysis of an edited Petri net graph. According to goals set by the user, the expert system performs the necessary reasoning about reachability and invariance properties of the analysed Petri net graph and displays inferred results to the user. The core of this expert system is based on rules capturing the general knowledge (i.e. expertise) about reachability and invariance of Petri nets. The definition of these rules forms an interesting and original part of the work. Prolog was used as the implementation language of the expert system.

KIMS: knowledge-based image management system

Abstract: A knowledge-based image management system called KIMS is developed for storage and retrieval of images. The aim of the system is to overcome some of the limitations inherent in Micro-IDBMS, a microcomputer-based image database management system. These limitations include abrupt failure and the inability to capture the concept of an image. Abrupt failure may occur during retrieval when there are no images that have the specified attribute value. The inability to capture the concept of an image is related to the general idea that an image is a representation. KIMS basically consists of an image knowledge base and an inference mechanism, in addition to Micro-IDBMS. The paper describes the role of image knowledge and presents the design and implementation details of KIMS.

Strategies for mixed reasoning with a graph-based reasoning engine

Abstract: An algorithm is presented that enables a novice to browse through a rule base and to interact with the rule base using a mixed-reasoning strategy. Essentially, the algorithm augments standard forward-reasoning inference with a technique for selecting goals whose antecedents are partially satisfied. The user is asked to confirm or deny the remaining antecedents of these rules. The novelty of the approach lies in the use of graphs as the nodes in combination with a powerful, graph-based inference engine, and in the heuristic for ordering and selecting goals for backward reasoning. The method presented is simple and general, and is easily adapted to any interactive knowledge-based system, owing to the domain independence of the algorithm.