Incorporating expertise into decision analysis based DSS

Artificial Intelligence/Expert Systems (AI/ES) techniques are proposed as means of combining the expertise of the experienced decision analyst with the computational methods of Decision Analysis (DA) to provide a user-friendly problem structuring, solution and querying environment which can be accessed directly by those responsible for managerial decisions.Such a system can be implemented as an Expert System shell. The key features of the shell are the inclusion of knowledge from previous problems through problem templates and the use of a combination of natural language and influence diagrams to communicate with users.This shell has the potential to further expand the use of DA by helping to overcome the DA skill shortage and to address some of the difficulties associated with purely AI/ES approaches to complex business problem solving. The greatest benefit of an Expert Decision Analyst is through integration with organisational DSS to support valuable information processing activities such as strategic decision-making.     

On becoming expert in the use of problem structuring methods

While much has been written about the theory and practice of using problem structuring methods (PSMs) there is little formal discussion of how individuals may effectively learn about their use. In this paper it is argued that a foundation for designing and providing a suitable learning environment lies in understanding the nature of being an expert user of PSMs and the knowledge that such experts hold and deploy. An analysis of how knowledge and expertise are grounded in the work involved in using PSMs is offered. This leads to some proposals for what is required if processes to support the acquisition of expertise in the use of PSMs are to be implemented.     

Distribution of Empty Railcars by an Expert System: A Case Study with Comparison with OR Approaches

The assignment of empty railcars by the SNCF (French Railways) distributors according to supply and demand from the regions could be considered as a conventional resource allocation problem. As such, it would be suitable for processing by integer programming. However, as the problem is currently solved by the reasoning process of the distributors, it involves much more strategic knowledge than linear programming. Based on this observation, we developed an allocation system centred on an expert system designed to reproduce the reasoning processes of the distributors. This paper describes the system, and the reasons why the ES approach seems more adapted to the problem than the OR pure solution.     

Guiding Data Analysts with Visual Statistical Strategies

Abstract

The concept of statistical strategy is introduced and used to develop a structured graphical user interface for guiding data analysis. The interface visually represents statistical strategies that are designed by expert data analysts to guide novices. The representation is an abstraction of the expert's concepts of the essence of a data analysis. We argue that an environment that visually guides and structures data analysis will improve data analysis productivity, accuracy, accessibility, and satisfaction in comparison to an environment without such aids, especially for novice data analysts. Our concepts are based on notions from cognitive science, and can be empirically evaluated. The interface consists of two interacting windows—the guidemap and the workmap. Each window contains a graph that has nodes and edges. The guidemap graph represents the statistical strategy for a specific statistical task (such as describing data). Nodes represent potential data analysis actions that can be taken by the system. Edges represent potential actions that can be taken by the analyst. The guidemap graph exists prior to the data analysis session, having been created by an expert. The workmap graph represents the complete history of all steps taken by the data analyst. It is constructed during the data analysis session as a result of the analyst's actions. Workmap nodes represent data sets, data models, or data analysis procedures that have been created or used by the analyst. Workmap edges represent the chronological sequence of the analyst's actions. One workmap node is highlighted to indicate which statistical object is the focus of the strategy. We illustrate our concepts with ViSta, the Visual Statistics system that we have developed.     

Simulation of ECB decisions and forecast of short term Euro rate with an adaptive fuzzy expert system

The elaboration of optimal monetary policy strategies, and the statistical estimation of monetary policy rules followed by European Central Bank (ECB) in the new currency area of the Euro, are difficult to follow with the standard statistical models. For this reason we have developed an adaptive fuzzy expert system in order to mimic the framework on which the monetary policy strategy of the ECB is based. The expert system knowledge base consists of a set of fuzzy and crisp rules located at two different hierarchical levels. The high-level of the system receives some intermediate output values from the low-level and processes this information by means of a set of crisp rules. The low-level prepares these intermediate output values with the use of a fuzzy inference engine applied to economic input variables. The use of an expert system allows for modelling the ECB behaviour with the use of wider scope of knowledge, when compared with more traditional computational techniques. Rules at different hierarchical levels and at different intra-level groups, allow for managing the potentially contradictory structure of the ECB strategy. The system has been tested on the economic and financial time series going from the January 1999 to September 2000. The system’s correct prediction was estimated to overall 70% and, considering the complexity of the task, the results obtained are promising.     

A fifth generation numerical modelling system in coastal zone

Nowadays, artificial intelligence (AI) technology is gradually integrated into the numerical modelling system to make the system more intelligent and more user-friendly. The characteristics of the fifth generation numerical modelling are connected with AI applications. The expert system technology as a widely applied AI technology is integrated into our modelling system for coastal water processes with traditional numerical computational tools and the data and graphical pre-processing and post-processing techniques. Five kinds of knowledge bases are built in the system to describe the existing expertise knowledge about model parameters, relations between parameters and physical conditions, various possible selections for parameters and rules of inference. The inference engine is designed to be driven by the confidence of correctness, and the rule base is built with the factor of confidence to link the various relations. The decision tree is designed to drive the inference engine to explore the route of selection procedure of modeling. The decision tree depends on the real problem specifications and can be modified during the dialogue between the system and the user. The forward chaining and backward chaining inference techniques are mixed together in the system to help matching the parameters in the model and the possible selections with sufficiently high confidence. The expert system technology is successfully integrated into the system to provide help for model parameter selection or model selection, and to make the numerical model system more accessible for non-expert users.     

A Comparison of Potential Applications of Expert Systems and Decision Support Systems

This paper describes ways in which decision support systems (DSS) and expert systems (ES) may be differentiated. It proposes a simplified systemic view which emphasises the functional differences of the environments within which each type of system is applied, thereby avoiding some of the problems often associated with attempting to define DSS and ES according to their constituent parts. This functional emphasis allows DSS and ES to be placed within a useful existing theoretical framework which takes account of contextual influences. The results of two independent surveys of DSS and ES applications in a sample domain-accountancy-are used to assess the view of their functionality held by users of the systems, particularly their functional differences. Some conclusions are drawn about the role of the user in the construction of such systems.     

智能的统计软件──统计专家系统

计算统计是数理统计的一个重要分支，而统计专家系统是计算统计中一极热门的话题。本文简介了统计专家系统的概念及发展过程，并根据本人所作的工作介绍了统计专家系统的基本结构和构造过程等。最后对未来的统计专家系统提出了一点设想。     

On Providing Quality of Service in Grid Computing through Multi-objective Swarm-Based Knowledge Acquisition in Fuzzy Schedulers

Nowadays, Grid computing is increasingly showing a service-oriented tendency and as a result, providing quality of service (QoS) has raised as a relevant issue in such highly dynamic and non-dedicated systems. In this sense, the role of scheduling strategies is critical and new proposals able to deal with the inherent uncertainty of the grid state are needed in a way that QoS can be offered. Fuzzy rule-based schedulers are emerging scheduling schemas in Grid computing based on the efficient management of grid resources imprecise state and expert knowledge application to achieve an efficient workload distribution. Given the diverse and usually conflicting nature of the scheduling optimization objectives in grids considering both users and administrators requirements, these strategies can benefit from multi-objective strategies in their knowledge acquisition process greatly. This work suggests the QoS provision in the grid scheduling level with fuzzy rule-based schedulers through multi-objective knowledge acquisition considering multiple optimization criteria. With this aim, a novel learning strategy for the evolution of fuzzy rules based on swarm intelligence, Knowledge Acquisition with a Swarm Intelligence Approach (KASIA) is adapted to the multi-objective evolution of an expert grid meta-scheduler founded on Pareto general optimization theory and its performance with respect to a well-known genetic strategy is analyzed. In addition, the fuzzy scheduler with multi-objective learning results are compared to those of classical scheduling strategies in Grid computing.     

Supporting a complex audit judgment task: An expert network approach

An auditor considers a tremendous amount of data when assessing the risk that the internal control (IC) structure of an entity will fail to prevent or detect significant misstatements in financial statements. The myriad of relationships between IC variables that must be identified, selected, and analyzed often makes assessing control risk a difficult task. While some general procedures and guidelines are provided, audit standards dictate no specifically set procedures and rules for making a preliminary control risk assessment (CRA). Rather, the procedures and rules are left mostly to auditor judgment. This paper considers the appropriateness of applying artificial intelligence (AI) techniques to support this audit judgment task. It details the construction of a prototype expert network; an integration of an expert system (ES) and a neural network (NN). The rules contained in the ES model basic CRA heuristics, thus allowing for efficient use of well-known control variable relationships. The NN provides a way to recognize patterns in the large number of control variable inter-relationships that even experienced auditors cannot express as a logical set of specific rules. The NN was trained using actual case decisions of practicing auditors.     

A numerical strategy to defectuous knowledge using

Knowledge-Based Systems are based on an often defectuous knowledge, be this knowledge acquired from experts or learned from examples.This paper presents a strategy designed to cope with defectuous knowledge: given a set of rules, it builds a similarity function over the work space of the problem. This similarity function together with a set of examples then enables case-based reasoning, through aK-nearest-neighbour-like process.Compared to other case-based reasoning techniques, the advantage of this approach is the following: the topology of the space is automatically induced from the given rules, instead of being explicitly provided (and tuned) by the expert.     

综合集成研讨厅的几个示例

顺应于国际系统反思的浪潮,以钱学森为首的我国系统科学家于1990年正式提出了从定性到定量综合集成方法论用以处理以往还原论所无法解决的开放复杂巨系统问题.1992年钱学森又继而提出了综合集成研讨厅的概念(简称为HWMSE)作为该方法论的实践平台,其中特别强调利用信息技术的优势.此后,不断有HWMSE的研制来体现综合集成方法论的功力,同样也引来质疑.通过3个未自称为HWMSE的示例,包括国外实验室反恐系统、笔者提出的“在线会场”及众所周知的搜索引擎,对每一个示例阐明了其中含有的构成研讨厅的专家体系、机器体系与知识体系的内容及在该研讨厅中正在进行的综合集成复杂问题求解的活动.期望以此扩展对HWMSE功能的理解,促进HWMSE的推广和综合集成方法论的实践.     

The Transparent and the Invisible in Professional Pedagogical Vision for Science Teaching

Science teacher educators use examples of practice to support teacher candidates (TCs) learning to engage in new forms of science teaching. However, interpretation of these examples assumes a level of expertise about practice TCs lack. This article describes a study designed to determine some of the differences between expert teachers' and TCs' professional pedagogical vision for science teaching. Specifically, the study examines what each group attends to (highlights) in examples of science teaching and how they interpret the events they attend to (codes). Both groups were asked to analyze video of classroom science teaching. Differences were found between TCs and expert teachers in terms of both highlighting and coding of science teaching practice. Four key areas of difference are described in detail: actor focus, questions, grain size, and enactment. The implications for science teacher education are discussed, in particular a set of tools and teacher education practices to support acculturating TCs into more a sophisticated professional pedagogical vision for science teaching. This article features a Research to Practice Companion Article . Please click on the supporting information link below to access.     

Is bushfire modelling a science? Impressions from recent legal cases involving bushfire damage

The southeast part of Australia is one of the more fire-prone areas of the world. Over the past decade a number of legal actions have been instituted in Australian courts to recover damage costs arising from various bushfires. Scientists with expertise in bushfire behaviour have been involved in several of these cases as expert witnesses.

In one case it was successfully argued that there was no “science of bushfire behaviour”, and therefore, expert evidence on bushfire behaviour was ruled inadmissable.

The legal concepts of expert evidence are summarized, as are the arguments concerning the scientific aspects of bushfire behaviour. An assessment is made of the current state of bushfire research as evidenced by the published literature. It was concluded that there are deficiencies in bushfire modelling at the present time, and that one of the reasons for this is the multidisciplinary nature of the problem. Some suggestions for future research are made to remedy these deficiencies.     

Searching for knowledge in a world flooded with facts

The wide availability of computer technology and large electronic storage media has led to an enormous proliferation of databases in almost every area of human endeavour. This naturally creates an intense demand for powerful methods and tools for data analysis. Current methods and tools are primarily oriented toward extracting numerical and statistical data characteristics. While such characteristics are very important and useful, they are often insufficient. A decision maker typically needs an interpretation of these findings, and this has to be done by a data analyst. With the growth in the amount and complexity of the data, making such interpretations is an increasingly difficult problem. As a potential solution, this paper advocates the development of methods for conceptual data analysis. Such methods aim at semi-automating the processes of determining high-level data interpretations, and discovering qualitative patterns in data. It is argued that these methods could be built on the basis of algorithms developed in the area of machine learning. An exemplary system utilizing such algorithms, INLEN, is discussed. The system integrates machine learning and statistical analysis techniques with database and expert system technologies. Selected capabilities of the system are illustrated by examples from implemented modules.     

Feedback stabilization for perturbed linear systems *

In this paper, we consider the questions of stabilization of perturbed (or uncertain) linear systems in Hilbert space. Perturbations of the system operator represent the uncertainty in the modelling process and could be bounded or even unbounded. Sufficient conditions are presented that guarantee stabilizability of the perturbed system given that the nominal (unperturbed) system is stabilizable. In particular it is shown that for certain class of perturbations weak and strong stabilizability properties are preserved for the same state feedback control. Two numerical examples are given to illustrate our theory.     

A Bayesian network approach to making inferences in causal maps

The main goal of this paper is to describe a new graphical structure called ‘Bayesian causal maps’ to represent and analyze domain knowledge of experts. A Bayesian causal map is a causal map, i.e., a network-based representation of an expert’s cognition. It is also a Bayesian network, i.e., a graphical representation of an expert’s knowledge based on probability theory. Bayesian causal maps enhance the capabilities of causal maps in many ways. We describe how the textual analysis procedure for constructing causal maps can be modified to construct Bayesian causal maps, and we illustrate it using a causal map of a marketing expert in the context of a product development decision.     

Applications of the theory of fuzzy sets

This survey summarizes the current state of the theory of fuzzy sets. The use of the theory in applications is illustrated with examples of control and modelling problems, as well as an application to the design of expert systems. Translated from Itogi Nauki i Tekhniki, Seriya Teoriya Veroyatnostei, Matematicheskaya Statistika, Teoreticheskaya Kibernetika, Vol. 29, pp. 83–151, 1990.     

MARKEX: An intelligent decision support system for product development decisions

A new methodology for the development of new products and an intelligent DSS, named MARKEX, which is an implementation of this methodology, are presented in this paper. The system acts as a consultant for marketers, providing visual support to enhance understanding and to overcome lack of expertise. The databases of the system are the results of consumer surveys, as well as financial information of the enterprises involved in the decision making process. The system's model base encompasses statistical analysis, preference analysis, and consumer choice models. MARKEX incorporates partial knowledge bases to support decision makers in different stages of the product development process.     

“I can name that Bayesian network in two matrixes!”

For a number of situations, a Bayesian network can be split into a core network consisting of a set of latent variables describing the status of a system, and a set of fragments relating the status variables to observable evidence that could be collected about the system state. This situation arises frequently in educational testing, where the status variables represent the student proficiency and the evidence models (graph fragments linking competency variables to observable outcomes) relate to assessment tasks that can be used to assess that proficiency. The traditional approach to knowledge engineering in this situation would be to maintain a library of fragments, where the graphical structure is specified using a graphical editor and then the probabilities are entered using a separate spreadsheet for each node. If many evidence model fragments employ the same design pattern, a lot of repetitive data entry is required. As the parameter values that determine the strength of the evidence can be buried on interior screens of an interface, it can be difficult for a design team to get an impression of the total evidence provided by a collection of evidence models for the system variables, and to identify holes in the data collection scheme. A Q-matrix - an incidence matrix whose rows represent observable outcomes from assessment tasks and whose columns represent competency variables - provides the graphical structure of the evidence models. The Q-matrix can be augmented to provide details of relationship strengths and provide a high level overview of the kind of evidence available. The relationships among the status variables can be represented with an inverse covariance matrix; this is particularly useful in models from the social sciences as often the domain experts’ knowledge about the system states comes from factor analyses and similar procedures that naturally produce covariance matrixes. The representation of the model using matrixes means that the bulk of the specification work can be done using a desktop spreadsheet program and does not require specialized software, facilitating collaboration with external experts. The design idea is illustrated with some examples from prior assessment design projects.