Building Knowledge-Based Systems -- Towards a Methodology

Dutch research on knowledge-based instructional systems: Introduction to the special issue

In Memoriam: Dr. Clive L. Dym (1942–2016)

Due to recent advances in knowledge-based systems technology it is suggested that students in a systems development course be given exposure to the concept of designing and implementing knowledge-based systems. The purpose of this paper is to describe the objective, scope, methods and procedures of knowledge-based projects. The first half of the paper introduces the knowledge-based approach and describes the structure and components of such a system. The second half of the paper deals with course design procedures. The scope, topics, project characteristics and experiences pertaining to such a project are covered here. A practical Knowledge-based Systems Development Life Cycle (KSDLC) is also proposed for use in such a course. It is a suitable methodology as it resembles the traditional Systems Development Life Cycle.

Artificial intelligence in pharmaceutical product formulation: knowledge-based and expert systems

Ever since MYCIN introduced the idea of computer-based explanations to the artificial intelligence community, it has come to be taken for granted that all knowledge-based systems (KBS) need to provide explanations. The first part of this dissertation proposes a cognitive learning theory based model that both clarifies the reasons as to why KBS need to provide explanations and serves as the basis for conceptualizing the provision of KBS explanations. Using the concepts of the feedforward and feedback operators of cognitive learning it develops strategies for providing KBS explanations and uses them to classify the various types of explanations found in current KBS applications. The roles of feedforward and feedback explanations within the context of the theory of cognitive skill acquisition and a model of expert judgment are also analyzed. These, together with past studies of KBS explanations, suggest that user expertise, the types of explanations provided, and the level of user agreement are significant factors that influence the explanation seeking behavior of users. The dissertation also explores the effects of the use of KBS explanations in judgmental decision making situations supported by a KBS. It identifies and considers four distinct categories of potential effects of the use of explanations--learning effects, perceived effects, behavioral effects, and effects on judgmental decision making. The second part of the dissertation empirically evaluates the explanation provision strategies in a laboratory experiment in which 80 novice and expert subjects used a KBS for financial analysis to make judgments under conditions of uncertainty. The experiment was designed specifically to investigate the following fundamental research questions: (1) To what extent are the various kinds of explanations used? (2) How does user expertise, the feedforward and feedback provision of explanations, and the level of user agreement influence the amount and the types of explanations that are used? and (3) Does the use of explanations affect the accuracy of judgmental decision-making and user perceptions of usefulness? Some of the major results relating to the determinants of the use of KBS explanations include: (1) user expertise is not a determinant of the proportion of explanations used but influences the types of explanations that are used, (2) explanation provision strategy is a critical determinant of the use of KBS explanations with feedback explanations being used significantly more than feedforward explanations, and (3) the three types of explanations are used in different proportions with the Why and How explanations being used significantly more that the Strategic explanations. (Abstract shortened by UMI.)

Ensuring reliability and enhancing quality of Knowledge Based Systems (KBS) are critical factors for their successful deployment in real-world applications. This is a broad task involving both methodological and formal approaches for designing rigorous Validation, Verification and Testing (VVT) methods and tools. Some of these can be adapted from conventional software engineering, while others rely on specific aspects of KBS.

The development, validation and implementation of knowledge-based systems

Artificial intelligence can assist in industrial control in many ways, but this paper concentrates on the role of knowledge-based systems in the areas of control, detection and diagnosis of faults, scheduling and planning. The paper argues that the application of knowledge-based techniques can provide a unification of these and other activities, and outlines the requirements of knowledge representation and inference techniques that are needed to make such systems a reality. The advances in hardware, including sensors as well as computers, that will be driven by and in turn stimulate the introduction of the knowledge-based industrial control systems are also considered.

Like other industrialized countries, France is currently enjoying a vogue for artificial intelligence and, generally, for hardware and software components and structures which will be needed for the design and implementation of the computer applications of the 1990s.Since public announcement of MITI's Fifth Generation Project in October 1981, the French scientific and industrial communications have exhibited increasing enthusiasm for AI languages, expert systems, man-computer interaction, novel computer architectures, and knowledge-based computer systems as a whole. The choice of the Prolog language for the Japanese project has stimulated many French industrialists to be aware of the existence of a basic AI tool designed mainly in France.In spite of the present fashion, often maintained by the journalistic milieu, it would be inaccurate to say that the French fifth generation project goes back to the Japanese announcement. The MITI project has certainly been a catalyst of ministerial and industrial awareness, but the bulk of ongoing projects stem from earlier work most often funded by government agencies.In spite of the current thrust in AI and the centralizing habit in France, a “flagship” AI project cannot be identified. French Research and Development initiatives in artificial intelligence in general, and expert systems in particular, correspond more to a set of distinct projects. These frequently complement each other in technical scope and in their scientific and industrial objectives.

The tools of artificial intelligence (AI) can be divided into two broad types: knowledge-based systems (KBSs) and computational intelligence (CI). KBSs use explicit representations of knowledge in the form of words and symbols. This explicit representation makes the knowledge more easily read and understood by a human than the numerically derived implicit models in computational intelligence. KBSs include techniques such as rule-based, modelbased, and case-based reasoning. They were among the first forms of investigation into AI and remain a major theme. Early research focused on specialist applications in areas such as chemistry, medicine, and computer hardware. These early successes generated great optimism in AI, but more broad-based representations of human intelligence have remained difficult to achieve (Hopgood, 2003; Hopgood, 2005).

The authors review and categorize the research in applications of artificial intelligence (AI) and expert systems (ES) in new product development (NPD) activities. A brief overview of NPD process and AI is presented. This is followed by a literature survey in regard to AI and ES applications in NPD, which revealed twenty four articles (twenty two applications) in the 1990–1997 period. The applications are categorized into five areas: expert decision support systems for NPD project evaluation, knowledge-based systems (KBS) for product and process design, KBS for QFD, AI support for conceptual design and AI support for group decision making in concurrent engineering. Brief review of each application is provided. The articles are also grouped by NPD stages and seven NPD core elements (competencies and abilities). Further research areas are pointed out.

The field of Soft Operational Research (Soft OR) has emerged from the attempt to address contextually rich, multi-actor ‘wicked’ problems that are not amenable to traditional ‘hard’ operational research techniques, which often rely on mathematical modelling. This study assesses the use of Soft OR techniques in climate change policymaking. Since climate change problems are classical wicked problems, many assume that Soft OR would be in high demand in developing climate change policy. And the review of the use of these techniques conducted here does find that in the cases where Soft OR methods have been used by academics and other non-governmental actors, they have consistently provided useful results for policymaking. It is puzzling therefore that there is little evidence of governments using Soft OR application in this area. We study two cases of explicit (in Bristol UK) and implicit use (Rhode Island, US) of such techniques by governments to explain why this is so. We argue that notwithstanding the challenges the two cases reveal in their application, Soft OR nevertheless has much to offer policymakers in the arena of climate change policymaking and deserve more attention and use.

This thesis aims to contribute to a theoretical understanding of the determinants and outcomes of the use of explanations provided in knowledge-based systems. It is important to deepen our understanding of explanation use, because explanations from knowledge-based systems are thought to be valuable by researchers, systems developers, and users. Moreover, considerable effort can be expended in the design and development of explanation facilities. Nevertheless, no accepted and well-tested theory could be found to account for the use of explanations. Previous empirical studies have resulted in some contradictory findings. The theoretical model proposed in this study is taken from cognitive learning theory, a cognitive effort perspective applied to the use of knowledge-based systems, and Toulmin's model of argumentation. The determinants of the frequency of explanation use investigated were (i) the level of expertise of the user, (ii) the goal of the user (learning or problem solving), and (iii) the nature of the problem solving situation (collaborative or non-collaborative). Collaborative situations are those in which both user and system can contribute knowledge to the problem solving process. The outcomes of explanation use investigated were (i) learning by the user, (ii) problem solving performance, and (iii) users' confidence in the system. Two experiments were performed in laboratory settings. In the first experiment, advanced-level students in a taxation subject used an operational system for capital gains tax calculations. This experiment focussed on the effect of the user's goal, whether learning or problem solving. In the second experiment, members of the general public used a knowledge-based system designed to assist with personal financial planning. This experiment focussed on the effect of the nature of the problem solving situation, whether collaborative or non-collaborative. Two versions of the knowledge-based system were used: one with explanations and one without. In both experiments, measures of frequency of explanation use were obtained from traces of program use. The results of the study are congruent with a cognitive effort perspective. This perspective implies that explanations will be used more when users perceive that the advantage to be gained from the explanations outweighs the cognitive effort in accessing the explanations. As predicted from this perspective, a goal of learning rather than problem solving led to higher frequency of use of explanations. Also as predicted, a requirement for collaborative problem solving led to higher frequency of use of explanations. This latter effect, however, was observed only with users classed as practiced users of explanations on the basis of their performance in familiarisation activities undertaken in preparation for the experiments. Results offered some support for cognitive learning theory. Increased frequency of use of explanations led to increased problem solving performance under some conditions. In the first study, with non-collaborative problem solving, the use of an Answer Help explanation was significantly associated with improved performance. In the second study, the frequency of use of explanations and Answer Help in total was positively related to problem solving performance. This situation was observed in situations where explanations were available. There was no significant difference in performance between groups with and without explanations. There was some evidence that the positive relationship between explanations and improved performance was more noticeable when problems of a collaborative nature were undertaken. The expected relationship between increased frequency of use of explanations and an increase in the amount of learning taking place was not observed. The experimental conditions were such that learning could have occurred in ways other than through use of explanations. In this light, results are not inconsistent with cognitive learning theory. The level of expertise of the user was found to be related to higher problem solving performance. The level of expertise of the user was found also to be related to greater use of explanations in familiarisation activities. Both these relationships were observed only in the first experiment, where the measure used for level of expertise related specifically to the problem domain of the knowledge-based system. Results did not support the propositions that were based on Touhnin's model of argumentation. Possibly, confidence in machine advice is associated with explanations that can provide the backing for arguments. In this research, however, this relationship was not observed. It appeared, rather, that the degree of confidence expressed in a system was associated with the nature of the system. Confidence ratings were significantly related to the level of expertise of individuals, but the direction of this relationship was different for the two different systems. With the system presented as fully operational, the more expert individuals gave the system a higher confidence rating. With the system presented with disclaimers because it was still under research, the more expert individuals gave the system a lower confidence rating.

This research is framed within the specific scope of the defense sector, addressing the expressed need of companies operating in this field to deploy dual use Artificial Intelligence (AI) technologies (military and civil) to amortize high development costs.Currently, AI technologies in this domain are developed solely based on problem-solving capabilities, operating under the specific exemption provided by the EU AI Act, and often disregarding current social demands.This renders the transfer or utilization of these technologies in the civil sphere, where compliance with current regulations and alignment with social expectations is imperative, virtually impossible.Consequently, companies are frequently compelled to develop distinct AI technologies depending on the application domain, preventing them from leveraging the advantages of dual-use development as intended.Therefore, the proposed challenge consists of designing a well-defined framework that enables defense companies to develop dual-use AI technologies.This challenge can only be successfully addressed by appropriately adopting Trustworthy AI principles throughout all processes of the AI System Development Life Cycle (SDLC).Only through this approach is it possible to obtain dual AI technologies with the desired capabilities that are both compliant with current legislation and aligned with societal demands.However, achieving this is not a trivial task; restrictions often conflict with model capabilities.There may be instances where only non-interpretable models (e.g., ANNs) can yield the expected results, yet, due to their practical application domain, they may violate established legal requirements or be deemed socially unacceptable.A review of the state of the art reveals that, while there are nascent works regarding the new EU AI regulation, there is a lack of research addressing the specific perspective proposed in this paper: the development of dual-use AI technologies for both defense and civil sectors.Thus, to address this challenge, the research aims to conduct research directed at establishing an Ecosystem of new knowledge.This will materialize as a standard for the design of a novel Dual AI System Development Life Cycle, incorporating the necessary set of associated development processes.The focus will be on obtaining Inherently Interpretable Models as well as Post-hoc Explainability Models, based on an AI-by-Design approach and under the Trustworthy AI paradigm.Implemented by companies, this framework will enable the development of dual AI technologies that are compliant with regulations and aligned with social demands.

Artificial intelligence (AI) has gained much popularity in various sectors and has found applications in multiple areas, including occupational health and safety (OHS) risk management of the high-risk construction, mining, and oil and gas sectors. OHS risk management centers on identifying, assessing and controlling occupational risks systematically to prevent work-related injuries, illnesses and deaths. This review presents the advances in AI applications for OHS risk management in these sectors and synthesizes their barriers for better application prospects. In the construction sector, AI can be employed in building information modeling during the design stage to identify and deal with the hazards of building models. AI can be deployed in construction sites through computer vision, sensor networks, knowledge-based systems, and machine learning to capture real-time site conditions, analyze the videos or pictures captured, and provide feedback to workers for appropriate responses. A similar setup involving the same components is also used for managing the OHS risks of surface or underground mining, particularly for monitoring the environmental conditions, detecting the presence of hazardous gases, and identifying hazards in locations that are remote and difficult to assess. Sensors can be attached to personal protective equipment and watches and the signals transmitted via Bluetooth to permit data collection for analysis and response by AI. In the oil and gas sector, sensors are extensively used to collect process safety data from wells, pipelines, valves, etc. for analytical and predictive Al. Al, especially, machine learning is used to create personalized training for workers based on their learning pace and characteristics. However, the major barriers identified are high cost, lack of support and skilled employees, ethical issues, and the uncertainty of AI.