Knowledge-based Explanations Research Articles

An Advanced Explainable Belief Rule-Based Framework to Predict the Energy Consumption of Buildings

The prediction of building energy consumption is beneficial to utility companies, users, and facility managers to reduce energy waste. However, due to various drawbacks of prediction algorithms, such as, non-transparent output, ad hoc explanation by post hoc tools, low accuracy, and the inability to deal with data uncertainties, such prediction has limited applicability in this domain. As a result, domain knowledge-based explainability with high accuracy is critical for making energy predictions trustworthy. Motivated by this, we propose an advanced explainable Belief Rule-Based Expert System (eBRBES) with domain knowledge-based explanations for the accurate prediction of energy consumption. We optimize BRBES’s parameters and structure to improve prediction accuracy while dealing with data uncertainties using its inference engine. To predict energy consumption, we take into account floor area, daylight, indoor occupancy, and building heating method. We also describe how a counterfactual output on energy consumption could have been achieved. Furthermore, we propose a novel Belief Rule-Based adaptive Balance Determination (BRBaBD) algorithm for determining the optimal balance between explainability and accuracy. To validate the proposed eBRBES framework, a case study based on Skellefteå, Sweden, is used. BRBaBD results show that our proposed eBRBES framework outperforms state-of-the-art machine learning algorithms in terms of optimal balance between explainability and accuracy by 85.08%.

Abstraction Hierarchy Based Explainable Artificial Intelligence

This work explores the application of Cognitive Work Analysis (CWA) in the context of Explainable Artificial Intelligence (XAI). We built an AI system using a loan evaluation data set and applied an XAI technique to obtain data-driven explanations for predictions. Using an Abstraction Hierarchy (AH), we generated domain knowledge-based explanations to accompany data-driven explanations. An online experiment was conducted to test the usefulness of AH-based explanations. Participants read financial profiles of loan applicants, the AI system’s loan approval/rejection decisions, and explanations that justify the decisions. Presence or absence of AH-based explanations was manipulated, and participants’ perceptions of the explanation quality was measured. The results showed that providing AH-based explanations helped participants learn about the loan evaluation process and improved the perceived quality of explanations. We conclude that a CWA approach can increase understandability when explaining the decisions made by AI systems.

The evolution of regional cross-border water regimes, the case of Deltarhine

In this study, we look at the evolution of a cooperative water regime in the delta of the Rhine catchment. In a Dutch–German case study, we focus on cross-border cooperation on the local and regional scale, describing and analyzing how a remarkably resilient and robust transboundary water regime has evolved over the course of 50 years. Context-, interest- and knowledge-based explanations contribute important insights into the evolution of the Deltarhine regime, and it is shown that the legal, institutional and socio-economic context shapes and constrains regional cross-border cooperation. Surprisingly in this regard, we find that European water directives have not yet played a decisive, catalyzing role for policy harmonization across borders. Finally, we show that key individuals play a crucial role in regime formation and development. We suggest that the presence of entrepreneurs and leaders adds explanatory power to current conceptual frameworks in international river basin management, thus meriting further research.

A Knowledge-Based Explanation of Franchise System Resources and Performance

The knowledge-based view (KBV) holds that the knowledge that an organization possesses is its most important strategic resource. Despite how important the adequate use of knowledge is to franchise system success, the KBV has yet to guide many franchising studies. The most important principle of KBV is to recognize system-wide resources through the eyes of knowledge-based resources. Using two separate secondary data sets, we examine the relationships between the knowledgeable employment of system resources and system performance. Empirical results reveal consistent support for our hypotheses. The ensuing discussion focuses on the implications of our findings.

Knowledge specialization and R&D collaboration

This paper contributes to the knowledge-based explanation of R&D networks. It argues that knowledge overlap and novelty are complementary inputs of a R&D alliance, in forms that depend upon the exploration breadth and depth of the R&D activity. The paper investigates how the hypothesis of specialization of the knowledge endowments can recover a number of characteristic empirical properties of a pattern of R&D collaboration in the economy. Implications for network evolution are discussed.

Project-conceptualisation in technological innovations: A knowledge-based perspective

Project management is widely regarded as a knowledge-based process. Critical to this process is the initial project-conceptualisation phase, especially in the context of technological innovations, but which has received little attention in the literature. Specifically, hardly any study has examined how project activities overlap, how project roles evolve, what skills are required and how much time is spent across different knowledge-based project-conceptualisation stages, or sets of activities. We explore these patterns and their knowledge-based explanations through a cross-case analysis of four technological innovations. Our contribution is a knowledge-based project-conceptualisation framework that deepens the appreciation of the evolving nature of the critical project management aspects across the highly uncertain project-conceptualisation phase. We then offer practical steps for project managers to effectively manage this project management phase.

Explanation Provision in Knowledge-Based Systems: A Theory-Driven Approach for Knowledge Transfer Designs

Knowledge-Based Systems (KBS) have been used in industry to free experts from mundane and routine decision making, to produce comparable and consistent decisions, and to retain the expertise of knowledgeable employees who may, for many reasons, leave a company. KBS are also desired to have the capacity to transfer knowledge to less-expert users of such systems. In this paper, Adaptive Character of Thought-Rational (ACT-R) theory is used as a foundation for the design of KBS explanations for the explicit purpose of facilitating knowledge transfer to the user. ACT-R (Anderson 1993) is a theory of cognitive skill acquisition that suggests a learner must first obtain certain facts about a new learning situation (declarative memory pieces) and then convert a series of facts into a set of rules that will produce accurate problem-solving skills (procedural memory pieces). Prior research has examined pieces of the ACT theory in its earlier forms, but no comprehensive tests examining the simultaneous effect of the multiple components have previously been completed. The current study addresses three questions based on ACT-R theory: (1) Can declarative-knowledge-based explanations improve declarative knowledge transfer? (2) Can declarative-knowledge-based explanations improve procedural knowledge transfer? (3) Can procedural-knowledge-based explanations improve procedural knowledge transfer? An experiment employing eight KBS, differing by types of KBS explanation prompts, which were designed to stimulate declarative and/or procedural knowledge transfer, was conducted with 294 accounting information systems students. An analysis of the results provides some support for the use of declarative-based KBS explanations for declarative knowledge transfer, strong support for the use of declarative-based KBS explanations for procedural knowledge transfer, but a lack of support for the use of procedural-based KBS explanations for procedural knowledge transfer. The results suggest that organizations may be able to improve the efficiency and effectiveness of training programs for knowledge workers through the application of KBS that include declarative knowledge-based explanations.

Confronting Nonregimes: Science and International Coral Reef Policy

Scientific communities and policy makers alike portray the worldwide degradation of coral reefs as a global issue that requires a policy response. Yet, an international legal regime on coral reefs is not on the global agenda, and no state considers it necessary. The absence of a regime is puzzling given a number of favorable factors that should facilitate the creation of a treaty. Moreover, existing theoretical explanations of international cooperation cannot adequately account for the case. This pioneer study on international coral reef management considers current policy-making activities in conjunction with information provided by multilateral scientific assessments. It offers a modified knowledge-based explanation of the case that relies on the distinction between different types of information. Missing knowledge about the transboundary effects of the problem impedes the formation of common interests and does not easily justify the need for a binding international agreement.

Using group support systems for developing a knowledge-based explanation facility

Group Support Systems (GSS) have been shown to improve the satisfaction and performance of multi-user decision making processes. This study describes how they can be used to acquire knowledge from multiple domain experts for building the explanation facilities of expert systems. In this study, a manual iterative Delphi approach was initially used to acquire knowledge from five experts to build the explanation facility of a financial analysis expert system. Next, GSS tools were utilized to repeat the acquisition of explanatory knowledge in a computer supported cooperative work environment. Comparative analysis suggests that the use of the parallel communication and group memory GSS tools enable process gains, such as getting more complete information, synergy, simulation, and learning, to be realized. Moreover, the fact that the domain experts worked together instead of separately reduced production blocking and cognitive inertia since they could build upon each other's comments. Significant reductions in the time required by experts and the knowledge engineer/facilitator suggests that GSS represent an efficient and effective approach for supporting the acquisition of explanatory knowledge from multiple experts.

Using ethnography in the design of an explanation system

In order for knowledge-based explanation systems to be acceptable, they must be useful and understandable to users. This implies that first, they should satisfy users' information needs and take account of their perspective; and second, they should be able to engage in dialogue with users. Much more progress has been made toward meeting the second condition than the first. Systems are still being produced that can engage in dialogue with users, but whose design reflects no systematic investigation of what people actually want (or need) to know about a given domain. Such lack of attention to the information needs of potential users is bound to limit the utility of any system. Because this issue relates to the concerns of social science as well as artificial intelligence, social scientists can help designers address it. One way of obtaining reliable data on the needs and characteristics of future users is ethnography, an anthropological method for gathering data in complex real-world settings. This article discusses some of the ways in which ethnography can contribute to the design process, drawing examples form an ongoing project to build an explanation system in migraine. Four aspects of our experience in using ethnography in the design process are discussed: rethinking basic design assumptions, investigating information needs, addressing the problem of perspective, and developing explanatory material. Based on this experience, the article concludes with the suggestion that the concept of explanation needs to be broadened still further to include more types of knowledge in the dialogue.

Problems with the Knowledge-Based Explanation of Memory and Development

This article is a critique of the claim that knowledge alone can account for the variance of memory and development. There are two parts: a brief historical sketch of the movement towards a knowledge-based view of memory and development and a discussion of some of its problems. The problems discussed are the equivocal nature of much of the data cited to support the knowledge-based view, the evidence that suggests that domain independent strategies influence memory and development, and the evidence for a domain independent tool of developmental import. Finally, the implications of these problems as they relate to theories of memory and development are discussed.

Problems with the Knowledge-Based Explanation of Memory and Development

This article is a critique of the claim that knowledge alone can account for the variance of memory and development. There are two parts: a brief historical sketch of the movement towards a knowledge-based view of memory and development and a discussion of some of its problems. The problems discussed are the equivocal nature of much of the data cited to support the knowledge-based view, the evidence that suggests that domain independent strategies influence memory and development, and the evidence for a domain independent tool of developmental import. Finally, the implications of these problems as they relate to theories of memory and development are discussed.

The partitioned support network for expert system justification

A knowledge representation used to capture support knowledge necessary for expert system justification, a kind of explanation, is discussed. A system is also presented that is being designed as a preliminary testbed for a technique of expert system explanation called knowledge-based explanation. Previous support knowledge representations in expert systems are reviewed and contrasted to the partitioned support network (PSN) introduced. PSNs are defined both in graphical and nongraphical notation. The significance and importance of the PSN knowledge representation are discussed and examples of its power are given. >