Expertise transfer and complex problems: using AQUINAS as a knowledge-acquisition workbench for knowledge-based systems

Boose, John H.; Bradshaw, Jeffrey M.

doi:10.1016/s0020-7373(87)80032-9

Cited by 241 publications

(44 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are KA tools based on this principle-for example, ETS (Boose 1986), KSS0 , and AQUINAS (Boose and Bradshaw 1987). Our work is similar in that we also view that some human knowledge is not readily available and explicit, and we aim at acquiring such knowledge in an intelligent manner.…”

Section: Related Workmentioning

confidence: 91%

“…Most KA tools require expert users to supply the description of a case or knowledge, outlining either completely or partially the attributes or concepts involved in the knowledge tasks or applications-for example, MORE (Kahn et al 1985a;1985b), AQUINAS (Boose et al 1995, Boose andBradshaw 1987), MIKE (Eisenstadt and Brayshaw 1990), SALT (Marcus 1987), KITTEN , KNACK (Klinker et al 1987), KSS0 , ALTO (Major and Reichgelt 1990), CMBKATs (Charlet et al 1992), KATEMES (Dieng et al 1992), CERISE (Vicat et al 1993), KADS (Schreiber et al 1993;van Heijst et al 1997), and many others . However, none of these KA tools has dealt with image-based knowledge ontology and devised a tool that facilitates either the acquisition or the refinement of the knowledge.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

ARKTOS: a knowledge engineering software tool for images

Soh¹,

Tsatsoulis²

2002

International Journal of Human-Computer Studies

View full text Add to dashboard Cite

-The goal of our ARKTOS project is to build an intelligent knowledge-based system to classify satellite sea ice images. It involves acquiring knowledge from sea ice experts, quantifying such knowledge as computational entities, and ultimately building an intelligent classifier. In this paper we describe a two-stage knowledge engineering approach that facilitates explicit knowledge transfer, converting implicit visual cues and cognition of the experts to explicit attributes and rules implemented by the engineers. First, there is a prototyping stage that involves interviewing sea ice experts, transcribing the sessions, identifying descriptors and rules, designing and implementing the knowledge, and delivering the prototype. The objective of this stage is to obtain a modestly accurate classification system quickly. Second, there is a refinement stage that involves evaluating the prototype, refining the knowledge base, modifying the design, and re-evaluating the improved system. Since the refinement is evaluation-driven, the experts and the engineers are motivated explicitly to improve the knowledge base and are able to communicate with each other using a common, consistent platform. Moreover, since the classification result is immediately available, both sides are able to efficiently assess the correctness of the system. To facilitate the knowledge engineering of the second stage, we have designed and built three Java-based graphical user interfaces: arktosGUI, arktosViewer, and arktosEditor. arktosGUI concentrates on feature-based refinement of specific attributes and rules. arktosViewer deals with regional evaluation. arktosEditor has a rule indexing and search mechanism and knowledge base editing capabilites. INTRODUCTIONThe goal of our ARKTOS (Advanced Reasoning using Knowledge for Typing Of Sea ice) project is to perform automated, intelligent satellite sea ice image classification. Our approach is to acquire the classification knowledge from sea ice geophysicists and photo-interpreters and implement the knowledge in a rule-based system that also utilizes image processing methodologies.To facilitate a knowledge transfer from the sea ice experts to the knowledge and software engineers, we employ a two-stage knowledge engineering approach that includes rapid prototyping and evaluation-driven refinement. The rapid prototyping allows the experts and the engineers to design and implement a modestly accurate classification system. Given the functional system, the experts are then able to evaluate its classification results. To facilitate the evaluation process, we create Java-based software tools such that implicit visual cues and cognition of the experts can be explicitly expressed and fine-tuned in attributes and rules. The prototyping stage involves

show abstract

Section: Related Workmentioning

confidence: 91%

Section: Related Workmentioning

confidence: 99%

ARKTOS: a knowledge engineering software tool for images

Soh¹,

Tsatsoulis²

2002

International Journal of Human-Computer Studies

View full text Add to dashboard Cite

show abstract

“…Similar to other 'contrived' knowledge elicitation techniques such as repertory grid and sorting techniques, it was originated in the domain of psychology [37]. It has been applied with increasing frequency in the field of knowledge acquisition and used as a 'technique' in its original form in predefined interviews as well as in the form of 'tools' for a computerized system [38,39]. As such, it appears that laddering has been developed to associate with more 'non-contrived' or 'natural' techniques such as interviewing and self-report [40].…”

Section: Related Workmentioning

confidence: 99%

A strategy for acquiring customer requirement patterns using laddering technique and ART2 neural network

Chen

Khoo

Yan

2002

Advanced Engineering Informatics

103

View full text Add to dashboard Cite

With the increasing interest and emphasis on customer demands in new product development, customer requirements elicitation (CRE) and evaluation have secured a crucial position in the early stage of product conceptualization. As such, it is highly desired to have a systematic strategy for soliciting customer requirements effectively, and subsequently, analyzing customer orientations quantitatively. For this purpose, a customer-oriented approach has accordingly been proposed and investigated in this study. It concerns both breadth and depth perspectives of customer requirements acquisition as well as customer and marketing analysis. This prototype system comprises two interrelated components, namely, the CRE and the customer/marketing analysis (CMA) modules. The process starts from the voice of customers and ends with the identified opportunities from marketing analysis. In the prototype system, the laddering technique has been employed to enable CRE via the so-called customer attributes hierarchy (CAM). In addition, the adaptive resonance theory, particularly ART2, neural network has been applied as a toolkit for further customer and marketing analysis. A case study on wood golf club design is used to demonstrate and validate this approach. q

show abstract

“…After giving concrete examples of topoi elicited thanks to actual interviews of experts, we will study thoroughly the exploitation of rating grids for topoi elicitation, as this technique is frequently used in knowledge acquisition tools such as ETS [5,6], KSS0 [43] Or AQUINAS [7,8].…”

Section: Knowledge Elicitation Techniquesmentioning

confidence: 99%

Acquisition of gradual knowledge

Dieng

Corby

Lapalut

1993

Knowledge Acquisition for Knowledge-Based Systems

View full text Add to dashboard Cite

Abstract. Topoi are gradual inference rules, often used by experts in several problem classes: they can be exploited at various phases of a knowledge-based system life cycle. They can be studied at the two levels distinguished by Newelh the knowledge level and the symbol level. Some knowledge elicitation techniques such as rating grids and some knowledge acquisition methods such as KADS and KOD can be exploited in order to facilitate the acquisition of topoi. At the symbol level, different representations and implementations of topoi can be proposed and topoi can be formalized through several qualitative physics formalisms. IntroductionThe notion of topoi 1 stems from Anscombre and Ducrot's argumentation theory [2]. A topos is a gradual inference rule such as "The more P, the more Q", "The less P, the less Q", ~'The more P, the less Q" or "The less P, the more Q", where P and Q are properties. The first two topoi are called positive topoi and the last two negative topoi. Several researchers studied the interest of such gradual knowledge for argumentation in linguistics [12,41] and for artificial intelligence: in particular, in the framework of knowledge-based systems (KBS), some researchers studied thoroughly the use of topoi for knowledge acquisition [23,27] Let us give a brief description of such research. In [12,41], the authors propose a topoi-based theory of argumentation in natural language. Topoi are seen as gradual rules that are applied to some elements of the signification of a phrase, and help to determine the argumentative orientation of the concerned utterance. A speaker chooses the topoi he supposes shared by his interlocutors. Topoi play an important role for representing expertise. In [41], the author distinguishes descriptive topoi and heuristic topoi and exploits the notion of topoi for explanations of an expert system. In [40], the author relies on the possibility theory and on fuzzy sets in order to reason on gradual rules. The dynamic gradual behaviour of a KBS is represented through a topoi base in [18]. The notion of conditional topoi is introduced in order to avoid conflicts between apparently contradictory topoi. The topoi base is exploited for specification and validation 1 The term "topoi" was chosen as a tribute to Aristotle's topoi.

show abstract

Expertise transfer and complex problems: using AQUINAS as a knowledge-acquisition workbench for knowledge-based systems

Cited by 241 publications

References 8 publications

ARKTOS: a knowledge engineering software tool for images

ARKTOS: a knowledge engineering software tool for images

A strategy for acquiring customer requirement patterns using laddering technique and ART2 neural network

Acquisition of gradual knowledge

Contact Info

Product

Resources

About