In the past few years, several large companies have published ethical principles of Artificial In... more In the past few years, several large companies have published ethical principles of Artificial Intelligence (AI). National governments, the European Commission, and inter-governmental organizations have come up with requirements to ensure the good use of AI. However, individual organizations that want to join this effort, are faced with many unsolved questions. This paper proposes guidelines for organizations committed to the responsible use of AI, but lack the required knowledge and experience. The guidelines consist of two parts: i) helping organizations to decide what principles to adopt, and ii) a methodology for implementing the principles in organizational processes. In case of future AI regulation, organizations following this approach will be well-prepared.
Popular press and media often make us believe that artificial intelligence technology is ethical ... more Popular press and media often make us believe that artificial intelligence technology is ethical or unethical by itself. In this paper, we will argue that organizations that develop or apply AI have certain choices they can make that will lead to a more or less responsible use of AI. By approaching those choices in a methodological way, organizations can make better decisions toward the ethical use of this powerful technology.
Constructing a planner for a particular application is a di cult job, for which little concrete s... more Constructing a planner for a particular application is a di cult job, for which little concrete support is currently available. The literature on planning is overwhelming and there is no clear synthesis of the various planning methods which could be used by knowledge engineers. In this paper, we show how a general, knowledge-level framework for conceptually specifying knowledgebased systems, can be of concrete use to support knowledge acquisition for planning systems. The framework encompasses three interrelated components: (1) problem-solving methods, (2) their assumptions and (3) domain knowledge. The presented analysis of planning performed in the framework can be considered as a library with reusable components, based on which planners can be con gured. Two experiments are presented that illustrate the use of the library in knowledge engineering.
Mostly, papers on problem-solving methods focus on the description of reasoning strategies and di... more Mostly, papers on problem-solving methods focus on the description of reasoning strategies and discuss their underlying assumptions as a side aspect. We take a complementary point of view and focus on these underlying assumptions as they play three important roles: first, assumptions are necessary to characterise the precise competence of a problem-solving method in terms of the tasks that can be solved by it, and in terms of the domain knowledge that is required by it. Second, assumptions are necessary to enable tractable problem solving for complex problems. Third, assumptions are necessary for appropriate interaction of the problem solver with its environment. Their introduction and refinement can be used to develop new problem-solving methods, or to adapt existing ones according to task and domain-specific circumstances of a given application. For this purpose, one requires a framework for dealing with these assumptions. This paper makes a step in this direction by summarising the assumptions that can be found in the literature on diagnosis with component-oriented device models. The main contribution of the paper is to collect these assumptions, to make their role in the reasoning process explicit, and to classify them.
Abstract . The IBROW, approach.,The proposed framework and language,provide,an effective organiza... more Abstract . The IBROW, approach.,The proposed framework and language,provide,an effective organization for constructing libraries with large horizontal cover, thus maximizing reusability and avoiding the brittleness of traditional, monolithic libraries. ,,,,,,, ,,,,
We expect reuse of parts of knowledge-based systems to become more and more important in the near... more We expect reuse of parts of knowledge-based systems to become more and more important in the near future, as opposed to building systems from scratch every time. It is, however, well known that there exists a tradeo between usability and reusability: the more components become reusable, the less usable they become. This is due to the fact that, in order to have a working system, a gap has to be bridged between the general reusable part and the speci c application, and the higher the reusability, the larger this gap. In this paper, we enhance the reusability of problem-solving methods by describing them in a task-neutral way. We ll the resulting gap by semi-automatically mapping the task-neutral method to a particular domain. in establishing this link. Most work on reuse has dealt with reuse in a task-speci c context. See for example , where reusable components are speci ed per task (e.g., planning, assessment, etc.), and , where an extensive library with PSMs for diagnosis is described. Recent work has, however, shown that PSMs for di erent tasks can be quite similar in spite of some apparent rst di erences van Heijst & Anjewierden, 1996). This raises the question whether it is possible to enlarge the reuse scope of problem-solving methods across di erent tasks. In this paper, we will investigate this question through so-called task-neutral problem-solving methods (TN-PSMs). TN-PSMs are context free methods in the sense that they do not use terminology that refers to a speci c task. However, it is well known that describing a reasoning process at a more general (i.e., more reusable) level, diminishes its applicability (i.e., usability) , because the distance between the reusable component and the domain knowledge at hand becomes larger, and thus establishing a mapping between them becomes harder. The aim of this paper is to remedy this tradeo between reusability and usability. Our solution consists of partially automating the establishment of the mapping between a PSM and the domain knowledge that it uses. Since we use a syntactical approach, a domain expert is needed for interpretation in case several mappings are found, and for explicating properties of the domain knowledge when necessary. The structure of this paper is as follows. In Section 2, we de ne the framework of this article. In Section 3, we brie y explicate the notion of domain ontology in the context of reuse. In Section 4, we introduce task-neutral PSMs, and in Section 5, we show how we can semi-automatically establish the mapping between the domain knowledge and the PSM. Section 6 points out related work and in Section 7, we present some conclusions. Throughout the paper, our approach is demonstrated with an example in the family domain.
... design process into four separate phases: (i) an analysis phase, (ii) a conceptual modeling .... more ... design process into four separate phases: (i) an analysis phase, (ii) a conceptual modeling ... which judges produce a different kind of knowled-ge through dogmatic legal categorizations it ... the assumption that their reasoning process follow some specific dogmatic patterns is not ...
A major bottleneck in business applications of Semantic Web technology is the lack of industry-st... more A major bottleneck in business applications of Semantic Web technology is the lack of industry-strength ontologies that go beyond academic prototypes. The design of such ontologies from scratch in a textbook-style ontology engineering process is in many cases unattractive, for it would require significant effort and because the resulting ontologies could not build on top of existing community commitment. Also, real-world problems of data and systems interoperability can only be overcome using Semantic Web technology if ontologies exist that represent the very standards currently in use in systems and databases. Of such standards, there exist at least four major categories:
Libraries with re-usable knowledge components are becoming increasingly important in Knowledge Ac... more Libraries with re-usable knowledge components are becoming increasingly important in Knowledge Acquisition. We propose a library of problem solving methods for diagnosis and describe some experiments and results concerning the usefulness of such a library for constructing and analyzing diagnostic strategies. A key notion is that each problem solving method is associated with suitability criteria, which are exploited in the process.
International Journal of Human Computer Studies, Apr 1, 1998
The support of knowledge acquisition on planning for engineering practical planning systems is di... more The support of knowledge acquisition on planning for engineering practical planning systems is discussed. There are two key themes within this support. The first theme illustrates the provision of task-oriented problem-solving templates to support the construction of planning systems. The second is the contrast of in how these templates are developed. These contracting approaches are called system-derived vs. domain-derived support.
In the past few years, several large companies have published ethical principles of Artificial In... more In the past few years, several large companies have published ethical principles of Artificial Intelligence (AI). National governments, the European Commission, and inter-governmental organizations have come up with requirements to ensure the good use of AI. However, individual organizations that want to join this effort, are faced with many unsolved questions. This paper proposes guidelines for organizations committed to the responsible use of AI, but lack the required knowledge and experience. The guidelines consist of two parts: i) helping organizations to decide what principles to adopt, and ii) a methodology for implementing the principles in organizational processes. In case of future AI regulation, organizations following this approach will be well-prepared.
Popular press and media often make us believe that artificial intelligence technology is ethical ... more Popular press and media often make us believe that artificial intelligence technology is ethical or unethical by itself. In this paper, we will argue that organizations that develop or apply AI have certain choices they can make that will lead to a more or less responsible use of AI. By approaching those choices in a methodological way, organizations can make better decisions toward the ethical use of this powerful technology.
Constructing a planner for a particular application is a di cult job, for which little concrete s... more Constructing a planner for a particular application is a di cult job, for which little concrete support is currently available. The literature on planning is overwhelming and there is no clear synthesis of the various planning methods which could be used by knowledge engineers. In this paper, we show how a general, knowledge-level framework for conceptually specifying knowledgebased systems, can be of concrete use to support knowledge acquisition for planning systems. The framework encompasses three interrelated components: (1) problem-solving methods, (2) their assumptions and (3) domain knowledge. The presented analysis of planning performed in the framework can be considered as a library with reusable components, based on which planners can be con gured. Two experiments are presented that illustrate the use of the library in knowledge engineering.
Mostly, papers on problem-solving methods focus on the description of reasoning strategies and di... more Mostly, papers on problem-solving methods focus on the description of reasoning strategies and discuss their underlying assumptions as a side aspect. We take a complementary point of view and focus on these underlying assumptions as they play three important roles: first, assumptions are necessary to characterise the precise competence of a problem-solving method in terms of the tasks that can be solved by it, and in terms of the domain knowledge that is required by it. Second, assumptions are necessary to enable tractable problem solving for complex problems. Third, assumptions are necessary for appropriate interaction of the problem solver with its environment. Their introduction and refinement can be used to develop new problem-solving methods, or to adapt existing ones according to task and domain-specific circumstances of a given application. For this purpose, one requires a framework for dealing with these assumptions. This paper makes a step in this direction by summarising the assumptions that can be found in the literature on diagnosis with component-oriented device models. The main contribution of the paper is to collect these assumptions, to make their role in the reasoning process explicit, and to classify them.
Abstract . The IBROW, approach.,The proposed framework and language,provide,an effective organiza... more Abstract . The IBROW, approach.,The proposed framework and language,provide,an effective organization for constructing libraries with large horizontal cover, thus maximizing reusability and avoiding the brittleness of traditional, monolithic libraries. ,,,,,,, ,,,,
We expect reuse of parts of knowledge-based systems to become more and more important in the near... more We expect reuse of parts of knowledge-based systems to become more and more important in the near future, as opposed to building systems from scratch every time. It is, however, well known that there exists a tradeo between usability and reusability: the more components become reusable, the less usable they become. This is due to the fact that, in order to have a working system, a gap has to be bridged between the general reusable part and the speci c application, and the higher the reusability, the larger this gap. In this paper, we enhance the reusability of problem-solving methods by describing them in a task-neutral way. We ll the resulting gap by semi-automatically mapping the task-neutral method to a particular domain. in establishing this link. Most work on reuse has dealt with reuse in a task-speci c context. See for example , where reusable components are speci ed per task (e.g., planning, assessment, etc.), and , where an extensive library with PSMs for diagnosis is described. Recent work has, however, shown that PSMs for di erent tasks can be quite similar in spite of some apparent rst di erences van Heijst & Anjewierden, 1996). This raises the question whether it is possible to enlarge the reuse scope of problem-solving methods across di erent tasks. In this paper, we will investigate this question through so-called task-neutral problem-solving methods (TN-PSMs). TN-PSMs are context free methods in the sense that they do not use terminology that refers to a speci c task. However, it is well known that describing a reasoning process at a more general (i.e., more reusable) level, diminishes its applicability (i.e., usability) , because the distance between the reusable component and the domain knowledge at hand becomes larger, and thus establishing a mapping between them becomes harder. The aim of this paper is to remedy this tradeo between reusability and usability. Our solution consists of partially automating the establishment of the mapping between a PSM and the domain knowledge that it uses. Since we use a syntactical approach, a domain expert is needed for interpretation in case several mappings are found, and for explicating properties of the domain knowledge when necessary. The structure of this paper is as follows. In Section 2, we de ne the framework of this article. In Section 3, we brie y explicate the notion of domain ontology in the context of reuse. In Section 4, we introduce task-neutral PSMs, and in Section 5, we show how we can semi-automatically establish the mapping between the domain knowledge and the PSM. Section 6 points out related work and in Section 7, we present some conclusions. Throughout the paper, our approach is demonstrated with an example in the family domain.
... design process into four separate phases: (i) an analysis phase, (ii) a conceptual modeling .... more ... design process into four separate phases: (i) an analysis phase, (ii) a conceptual modeling ... which judges produce a different kind of knowled-ge through dogmatic legal categorizations it ... the assumption that their reasoning process follow some specific dogmatic patterns is not ...
A major bottleneck in business applications of Semantic Web technology is the lack of industry-st... more A major bottleneck in business applications of Semantic Web technology is the lack of industry-strength ontologies that go beyond academic prototypes. The design of such ontologies from scratch in a textbook-style ontology engineering process is in many cases unattractive, for it would require significant effort and because the resulting ontologies could not build on top of existing community commitment. Also, real-world problems of data and systems interoperability can only be overcome using Semantic Web technology if ontologies exist that represent the very standards currently in use in systems and databases. Of such standards, there exist at least four major categories:
Libraries with re-usable knowledge components are becoming increasingly important in Knowledge Ac... more Libraries with re-usable knowledge components are becoming increasingly important in Knowledge Acquisition. We propose a library of problem solving methods for diagnosis and describe some experiments and results concerning the usefulness of such a library for constructing and analyzing diagnostic strategies. A key notion is that each problem solving method is associated with suitability criteria, which are exploited in the process.
International Journal of Human Computer Studies, Apr 1, 1998
The support of knowledge acquisition on planning for engineering practical planning systems is di... more The support of knowledge acquisition on planning for engineering practical planning systems is discussed. There are two key themes within this support. The first theme illustrates the provision of task-oriented problem-solving templates to support the construction of planning systems. The second is the contrast of in how these templates are developed. These contracting approaches are called system-derived vs. domain-derived support.
Uploads
Papers by V. Richard Benjamins