Stand-alone Expert Systems Research Articles

A HYBRID BAYESIAN NETWORK-BASED MULTI-AGENT SYSTEM AND A DISTRIBUTED SYSTEMS ARCHITECTURE FOR THE DRUG CRIME KNOWLEDGE MANAGEMENT

In this paper, we describe an approach for building a hybrid Bayesian network-based multi-agent system for drug crime knowledge management. We use distributed artificial intelligence architecture to create a multi-agent information system that integrates distributed knowledge sources and information to aid decision-making. Our comparison of the hybrid system with a previously developed stand-alone expert system Sherpa, which was in use at a large drug crime investigation facility, shows that the current system compares similar to the existing system in terms of efficiency and effectiveness of knowledge management. We illustrate how the proposed hybrid bayesian network-based can be implemented in the distributed computing network environment.

Information sharing between heterogeneous uncertain reasoning models in a multi-agent environment: a case study

The issue of information sharing and exchanging is one of the most important issues in the areas of artificial intelligence and knowledge-based systems (KBSs), or even in the broader areas of computer and information technology. This paper deals with a special case of this issue by carrying out a case study of information sharing between two well-known heterogeneous uncertain reasoning models: the certainty factor model and the subjective Bayesian method. More precisely, this paper discovers a family of exactly isomorphic transformations between these two uncertain reasoning models. More interestingly, among isomorphic transformation functions in this family, different ones can handle different degrees to which a domain expert is positive or negative when performing such a transformation task. The direct motivation of the investigation lies in a realistic consideration. In the past, expert systems exploited mainly these two models to deal with uncertainties. In other words, a lot of stand-alone expert systems which use the two uncertain reasoning models are available. If there is a reasonable transformation mechanism between these two uncertain reasoning models, we can use the Internet to couple these pre-existing expert systems together so that the integrated systems are able to exchange and share useful information with each other, thereby improving their performance through cooperation. Also, the issue of transformation between heterogeneous uncertain reasoning models is significant in the research area of multi-agent systems because different agents in a multi-agent system could employ different expert systems with heterogeneous uncertain reasonings for their action selections and the information sharing and exchanging is unavoidable between different agents. In addition, we make clear the relationship between the certainty factor model and probability theory.

A cooperative agent modelling approach for process planning

A well designed computer-aided process planning (CAPP) system bridges the gap between CAD and CAM. A number of systems have recently been developed relying on a standalone expert system. However, because of over-complexity, many such systems cannot be effectively applied to industrial enterprises in practice. Moreover, the modern computer integrated manufacturing system (CIM) requires the CAPP system to be extendible and flexible for practical industrial applications. It is hardly possible to develop the extensive industrial CAPP system by using only one large expert system. To overcome these weaknesses, a new cooperative agent model is presented for process planning in this paper that satisfies five major requirements: Autonomy, Flexibility, Interoperability, Modularity and Scalability. In accordance with this framework proposed, a machining cooperative process planning system (Machining CoCAPP) is specifically developed for demonstration purpose. The system modelling, agent structure design, cooperation and coordination mechanism, and case study of the Machining CoCAPP are presented.

Isomorphic transformations of uncertainties for incorporating EMYCIN-style and PROSPECTOR-style systems into a distributed expert system

In the past, expert systems exploited mainly the EMYCIN model and the PROSPECTOR model to deal with uncertainties. In other words, a lot of stand-alone expert systems which use these two models are available. If we can use the Internet to couple them together, their performance will be improved through cooperation. This is because the problem-solving ability of expert systems is greatly improved by the way of cooperation among different expert systems in a distributed expert system. Cooperation between different expert systems with these two heterogeneous uncertain reasoning models is essentially based on the transformations of uncertainties of propositions between these two models. In this paper, we discovered the exactly isomorphic transformations uncertainties between uncertain reasoning models, as used by EMYCIN and PROSPECTOR.

A cooperative framework for process planning

Tremendous efforts have been made to develop computer-aided process planning (CAPP) systems. However, most of the systems are based on standalone expert systems which are hard to extend and are inflexible for practical industrial applications. This paper introduces a cooperative process planning system named CoCAPP ( Cooperative Computer-Aided Process Planning) which has been developed by the authors, utilizing cooperative agent technology. The system is based on the five requirements: Autonomy, Flexibility, Interoperability, Modularity and Scalability, in order to minimize currently available system's weakness and improve applicable characteristics. The framework, the modeling of agents, and the cooperative planning mechanism are detailed. A case study of a machining CoCAPP system is also introduced to illustrate the implementation of CoCAPP concept.

Eighth Workshop on the Validation and Verification of Knowledge-Based Systems

The Workshop on the Validation and Verification of Knowledge-Based Systems gathers researchers from government, industry, and academia to present the most recent information about this important development aspect of knowledge-based systems (KBSs). The 1995 workshop focused on nontraditional KBSs that are developed using more than just the simple rule-based paradigm. This new focus showed how researchers are adjusting to the shift in KBS technology from stand-alone rule-based expert systems to embedded systems that use object-oriented technology, uncertainty, and nonmonotonic reasoning.

Expert Systems and Analytical Chemistry: Recent Progress in the ACexpert Project

In this paper we discuss the application of expert system technology in the analytical chemistry domain and summarizes the progress made in the implementation of the ACexpert project. We start with a brief description of the software tools that are commonly used in construction of expert system applications as well as the inference mechanisms employed to search the knowledge base. A brief review of the achievements made world wide during the past five years in the development of expert systems for use in the chemistry domain follows. This paper particularly addresses the difficulties found in the knowledge acquisition process. An effective knowledge encoding scheme, named the knowledge domain matrix process (KDM), that has been developed in our laboratory as an knowledge encoding alternative, is described. In the final section, we describe the development of several stand-alone expert system modules based on implementation of the KDM scheme within the framework of the ACexpert project. The importance of t...

Rule generation as an alternative to knowledge acquisition: A systems architecture for medical informatics

This paper describes a clinical support systems workbench, DiagaiD, based on an efficient transfer of patient data between health care professionals and clinical subsystems. The DiagaiD workbench provides tools for decision support developments for open-loop systems. A leading ambition for our development work has been to establish a data analysis and knowledge elicitation workbench, in which medical professionals entirely by themselves can create stand-alone expert systems.

Transforming standalone expert systems into a community of cooperating agents

Distributed Artificial Intelligence (DAI) systems in which multiple problem-solving agents cooperate to achieve a common objective is a rapidly emerging and promising technology. However, as yet, there have been relatively few reported cases of such systems being employed to tackle real-world problems in realistic domains. One of the reasons for this is that DAI researchers have given virtually no consideration to the process of incorporating pre-existing systems into a community of cooperating agents. Yet reuse is a primary consideration for any organisation with a large software base. To redress the balance, this paper reports on an experiment undertaken at the CERN laboratories in which two pre-existing and standalone expert systems for diagnosing faults in a particle accelerator were transformed into a community of cooperating agents. The experiences and insights gained during this process provide a valuable first step towards satisfying the needs of potential users of DAI technology—identifying the types of changes required for cooperative problem solving, quantifying the effort involved in transforming standalone systems to ones suitable for cooperation and highlighting the benefits of a cooperating systems approach in a realistic industrial application.

The reliability of embedded AI systems

An artificial intelligence (AI) software reliability model, which accounts for the trade-off between correctness and response time, fuzzy rather than binary output, faults in fundamental techniques used in AI programs, and the evolving database that AI programs rely on, is discussed. The model can be reduced to a simpler form for most applications, such as stand-alone expert systems and game-playing programs. In its full generality, it is applicable to real-time planning systems involving hardware and software.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Expert systems in management and administration - are they really different from Decision Support Systems?

Historically, the fields of Decision Support Systems and Expert Systems grew up from different backgrounds; Management Information Systems and Artificial Intelligence, respectively. As a result, many people still regard them as unrelated fields, although Operational Researchers are working in both of them. Expert Systems in particular have found it difficult to escape from an image of being isolated; not just from Decision Support Systems, but from all other types of computer system. Succesful applications appeared to be confined to very well-defined tasks; either small systems for tasks of an operational nature, or large systems requiring expensive specialised computer support. The spread of Expert Systems into management and administrative applications from the scientific/technical domains of the early systems has been very slow. The view put forward in this paper is that for Expert Systems to be applied to problems in management or administration, the traditional ‘closed-world’ picture of an Expert System is usually inadequate. The real manifestation of the Expert System's role (and indeed that of the human expert) involves much more negotiation and interaction than in scientific/technical domains. The consequences for the resulting system are that it looks much more like the traditional picture of a Decision Support System than a classic standalone Expert System. The argument will be illustrated by reference to examples of ‘Expert System’ developments, including systems for production control, crime pattern analysis, telephone marketing, and the drawing up of contracts.

Feasibility study for the construction of an integrated expert system in high-performance liquid chromatography

An integrated expert system consisting of several stand-alone expert systems was developed to assist the chromatographer in the determination of optimum high-performance liquid chromatographic conditions, i.e., after a good “first guess”, an elution within a reasonable analysis time and with adequate resolution. The implementation and linking of the systems were performed by means of the expert system building tool KES. The knowledge incorporated in this expert system is described.

Integrating expert systems for high-performance liquid chromatographic method development

A complex expert system for high-performance liquid chromatographic method development has been obtained by linking several smaller stand-alone expert systems (modules), each representing a part of the entire domain. The modules LABEL, DASH and LIT select initial chromatographic conditions (first guess), the modules LABEL′, DASH′ and LIT′ perform retention optimization and the module SLOPES, which consists of the stand-alone expert systems VARIABLES, DESIRE and CRISE, is used for selectivity optimization. The linking is achieved through a supervisor, that is able to decide which of the former stand-alone expert systems must be activated at each stage of the development. The stand-alone expert systems were implemented by the use of the expert system building tool (ESBT) KES. To allow communication between the modules, they were embedded in C-language programs, without modification of the knowledge of the smaller systems.

An application of expert systems in geotechnical risk assessment for surface coal mine design

ABSTRACT This paper describes the on-going development of an advanced computer system for the assessment of geotechnical risk in surface coal mines. In particular it is concerned with detailing the practical problems and solutions involved in linking computerised geological models, specialist mining software packages arid expert systems when applied to slope stability prediction. The development of ESDS, an expert system for slope stability assessment, is reviewed and the knowledgebase structures described. The limitations of stand-alone expert systems are discussed and the paper concentrates on the computing problems involved in developing an integrated system which accesses a series of mining software packages. The interfacing of geological and design databases for a project and ESDS is presented and the problems of designing a decision support system around the available data are illustrated. The paper concludes by presenting an example which illustrates how ESDS may be utilised as a decision support system at the design stage of a UK surface coal mine.

Network Restoration Expert System

Despite preventive measures, system breakdowns cannot entirely be avoided. In the attempt to minimize the time of outage, the use of intelligent restoration strategies adapted to the special network and breakdown case, plays an important role. For that aim, an expert system will provide a reasonable tool in finding the appropriate proceedings. This tool in principle can be used in real time operation as well as in training the operational staff. The paper describes experiences gained with a (presently) stand-alone expert system for network restoration and its extension to obtain a training simulator for restorative strategies.

CEST: An Expert System Function Library and Workbench forUnix® System/C Language

Integrating expert system components into production software can be difficult, because environments for developing expert systems typically are not compatible with traditional software-engineering technology. To deal with this problem, we are developing CEST, a C-language expert system toolset. It is a library of inference engines — implemented as C functions that can be called from C programs — and a workbench of knowledge-engineering support tools. CEST allows easy integration of expert system components into C-based software systems, and provides knowledge-engineering support tools analogous to traditional software-engineering support tools. The first tool written for CEST is AVIEN, a backward-chaining attribute-value inference engine. It has been widely distributed within AT&T, and has been used to build both stand-alone expert systems and C-based hybrid systems. In particular, the Quality Assurance Center at AT&T Bell Laboratories is using AVIEN in software tools being developed for quality and reliability analysis.

Artificial intelligence costs, benefits, and risks for selected spacecraft ground system automation scenarios

In response to a number of high-level strategy studies in the early 1980s, Expert Systems and Artificial Intelligence (AI/ES) efforts for spacecraft ground systems have proliferated in the past several years primarily as individual small to medium scale applications. It is useful to stop and assess the impact of this technology in view of lessons learned to date and, hopefully, to determine if the overall strategies of some of the earlier studies both are being followed and still seem relevant. To achieve that end four idealized ground system automation scenarios and their attendant AI architectures are postulated and benefits, risks, and lessons learned are examined and compared. These architectures encompass: (1) no AI (baseline), (2) standalone expert systems, (3) standardized, reusable knowledge base management systems (KBMS), and (4) a futuristic unattended automation scenario. The resulting artificial intelligence lessons learned, benefits, and risks for spacecraft ground system automation scenarios are described.