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

doi:10.1016/S0020-7373(87)80032-9

International Journal of Man-Machine Studies

Volume 26, Issue 1, January 1987, Pages 3-28

https://doi.org/10.1016/S0020-7373(87)80032-9 Get rights and content

Abstract

Acquiring knowledge from a human expert is a major problem when building a knowledge-based system. Aquinas, an expanded version of the Expertise Transfer System (ETS), is a knowledge-acquisition workbench that combines ideas from psychology and knowledge-based systems research to support knowledge-acquisition tasks. These tasks include eliciting distinctions, decomposing problems, combining uncertain information, incremental testing, integration of data types, automatic expansion and refinement of the knowledge base, use of multiple sources of knowledge and providing process guidance. Aquinas interviews experts and helps them analyse, test, and refine the knowledge base. Expertise from multiple experts or other knowledge sources can be represented and used separately or combined. Results from user consultations are derived from information propagated through hierarchies. Aquinas delivers knowledge by creating knowledge bases for several different expert-system shells. Help is given to the expert by a dialog manager that embodies knowledge-acquisition heuristics.

Aquinas contains many techniques and tools for knowledge acquisition; the techniques combine to make it a powerful testbed for rapidly prototyping portions of many kinds of complex knowledge-based systems.

References (58)

J.H. Boose
A knowledge acquisition program for expert systems based on personal construct psychology
International Journal of Man-Machine Studies
(1985)
G. Shafer et al.
Languages and designs for probability judgment
Cognitive Science
(1985)
J. Adams
Probabilistic reasoning and certainty factors
M. Alpert et al.
A progress report on the training of probability assessors
D. Antonelli
The application of artificial intelligence to a maintenance and diagnostic information system (MDIS)
M. Benda et al.
Boeing blackboard system
Benda, M., Jagannathan, V. & Dodhiawala, R., On optimal cooperation of knowledge sources. Workshop on Distributed...
R. Beyth-Marom et al.
J.H. Boose
Personal construct theory and the transfer of human expertise
J.H. Boose

J.H. Boose

Rapid acquisition and combination of knowledge from multiple experts in the same domain

D.E. Brown

Expert systems for design problem-solving using design refinement with plan selection and redesign

T. Bylander et al.

Generic tasks in knowledge-based reasoning: the “Right,” level of abstraction for knowledge acquisition

T. Bylander et al.

CSRL: a language for classificatory problem solving and uncertainty handling

AI Magazine

(1986)

B. Chandrasekaran

Generic tasks in knowledge-based reasoning: high-level building blocks for expert system design

IEEE Expert

(1986)

P. Cheeseman

In defense of probability

W. Clancey

Heuristic classification

C.H. Coombs et al.

L. Eshelman et al.

MOLE: a tenacious knowledge-acquisition tool

B.R. Gaines

An overview of knowledge acquisition and transfer

International Journal of Man-Machine Studies

(1987)

B.R. Gaines et al.

New directions in the analysis and interactive elicitation of personal construct systems

B.R. Gaines et al.

Induction of inference rules for expert systems

Fuzzy Sets and Systems

(1985)

J. Gordon et al.

The Dempster-Shafer theory of evidence

T. Gruber et al.

Design for acquisition principles of knowledge system design to facilitate knowledge acquisition

International Journal of Man-Machine Studies

(1987)

M. Henrion

Propagating uncertainty by logic sampling in Bayes' networks

D.N. Hinkle

The change of personal constructs from the viewpoint of a theory of implications

R.A. Howard et al.

Influence diagrams

R. Hurley

G. Kahn et al.

MORE: an intelligent knowledge acquisition tool

Cited by (164)

Developing a Capability to Elicit and Structure Psychosocial Decision Information within Computational Models
2015, Procedia Manufacturing
There is a recognized need to develop computational models that can represent and simulate the decision making process of various groups across socio-cultural domains [5]. Yet, developing such models can be greatly hampered by the need to acquire and represent information pertaining to the psychological and social aspects of decision-making within these groups. Currently, there are numerous techniques and tools to help facilitate the elicitation and structuring of knowledge within expert-type systems—particularly those that focus on technical processes such as mechanical troubleshooting [3]. However, few techniques and tools have been developed for models that are intended to represent and assess the decision making of groups within different societies—particularly including cultural elements within these societies. This paper seeks to help address this challenge by discussing an approach to eliciting and structuring cross-cultural psychosocial and behavioral-economic elements within a theory-based assessment model. This work was developed to address the needs of Sandia National Laboratories’ Behavioral Influence Assessment modeling capability, which assesses decision-making within societies. The main component of the knowledge engineering effort is what we call the “knowledge structure.” The knowledge structure acts as scaffolding for the organization of psychosocial processes underlying decision-making, as well as the actual content of that knowledge with respect to a modeled society.
A questioning based method to automatically acquire expert assembly diagnostic knowledge
2014, CAD Computer Aided Design
Citation Excerpt :
Similarly, the KBDT system [23] takes in an initial set of vocabulary terms, acronyms and euphemisms, and works with examples to elicit design rationale. An early system ACQUINAS [24] also looks at knowledge for configuring systems using rating grids, and again uses examples—but this is applicable for domains with a prior, clear, domain model. Another research effort that targets similar objectives uses Fuzzy Cognitive Models [25] as the framework for modeling diagnostic knowledge.
In the domain of manual mechanical assembly, expert knowledge is an important means of supporting assembly planning that leads to fewer issues during actual assembly. Knowledge based systems can be used to provide assembly planners with expert knowledge as advice. However, acquisition of knowledge remains a difficult task to automate, while manual acquisition is tedious, time-consuming, and requires engagement of knowledge engineers with specialist knowledge to understand and translate expert knowledge. This paper describes the development, implementation and preliminary evaluation of a method that asks a series of questions to an expert, so as to automatically acquire necessary diagnostic and remedial knowledge as rules for use in a knowledge based system for advising assembly planners diagnose and resolve issues. The method, called a questioning procedure, organizes its questions around an assembly situation which it presents to the expert as the context, and adapts its questions based on the answers it receives from the expert.
User expertise in contemporary information systems: Conceptualization, measurement and application
2013, Information and Management
Citation Excerpt :
In addition, numerous studies have identified that knowledge sharing is an essential part of knowledge management, and it is imperative to ES success [e.g. 99, 100]. Studies of several disciplines; IS [e.g. 101, 102], business [e.g. 103], and psychology [e.g. 104, 105, 106]; suggest that experts, are willing, able and motivated to convey their superior knowledge and skills to novices. The test of the structural model includes estimates of the path coefficient, which indicate the strength of the relationship between the independent and dependent variable, and the R2 values, which represent the amount of variance explained by the independent variable/s. Together, the R2 and the path coefficient (loadings and significance) indicate how well the data supports the hypothesized model [91].
The development of user expertise is a strategic imperative for organizations in hyper-competitive markets. This paper conceptualizes opreationalises and validates user expertise in contemporary Information Systems (IS) as a formative, multidimensional index. Such a validated and widely accepted index would facilitate progression of past research on user competence and efficacy of IS to complex contemporary IS, while at the same time providing a benchmark for organizations to track their user expertise. The validation involved three separate studies, including exploratory and confirmatory phases, using data from 244 respondents.
From knowledge science to symbiosis science
2013, International Journal of Human Computer Studies
Citation Excerpt :
In 1985, Brian Gaines produced an early version of a diagram predicting the future of knowledge systems, and showing Wisdom as the pinnacle of that evolution (Fig. 1). Four years later, I was honored to join Brian, along with my mentor and friend John Boose (Boose, 1986; Boose and Bradshaw, 1987; Bradshaw and Boose, 1990; Bradshaw et al., 1991; Shema et al., 1990), in sending out a call for papers for a Workshop on Wisdom-Based Systems that was to take place on June 22–24, 1989 at the Rosario Resort on Orcas Island in Washington state. Here is a paragraph from that call:
In the mid-1980s, Brian Gaines first developed a model to predict the trajectory of progress in human–computer relationships, including how the knowledge science research programme would naturally transform itself over time into something he called “symbiosis science.” In this article, we reflect both on the extraordinary prescience of this model, and the contributions and challenges faced by researchers intent on progressive achievement toward the aspirations it inspires.
Failure analysis expert system for onshore pipelines. Part - I: Structured database and knowledge acquisition
2011, Expert Systems with Applications
Citation Excerpt :
These techniques have been developed to help DE the reason about his expertise, and the KE to understand it. Some of the techniques mainly for data extraction are: Repertory Grid Analysis (RGA); (Boose, 1985; Gaines & Shaw, 1980), and its derivations, such as VODKA (Tseng & Lin, 2009); KAMET (Chu & Hwang, 2008), ETS (Boose, 1985), AQUINAS (Boose & Bradshaw, 1986); KITTEN (Shaw & Gaines, 1987) and others. Machine Learning Approach: In this case, the knowledge approximation from domain examples provided by the experts and using software with learning algorithms, for example ID3 (Quinlan, 1986) and AQ11 (Michalski & Chilausky, 1980) which are two well-known commercial learning algorithms.
In this article is described a knowledge-based system or expert system for failures identification in onshore pipelines. This expert system is called Failure Analysis Expert System (FAES). The FAES development has been split in two parts. In the present part I, the database structure and knowledge acquisition process are described, while in second part, the End-User interface and learning algorithm will be described.
The proposed FAES includes a structured database with document processing of typical failures of pipeline collected from failure analysis reports and which were supported by expertise of failure analysis experts. A total de 854 cases of onshore pipeline failures were considered for FAES development; 683 cases for training and 171 cases for testing. Several failure mechanisms were identified with the following frequency order: external corrosion, internal corrosion, third parties, erosion, material failure, and vandalism. For machine learning, an inductive learning algorithm through Artificial Neural Network (ANN) was used.
VODKA: Variant objects discovering knowledge acquisition
2009, Expert Systems with Applications
Many knowledge acquisition methodologies have been proposed to elicit rules systematically with embedded meaning from domain experts. But, none of these methods discusses the issue of discovering new modified objects in a traditional classification knowledge based system. For experts to sense the occurrence of new variants and revise the original rule base, to collect sufficient relevant information becomes increasingly important in the knowledge acquisition field. In this paper, the method variant objects discovering knowledge acquisition (VODKA) we proposed includes three stages (log collecting stage, knowledge learning stage, and knowledge polishing stage) to facilitate the acquisition of new inference rules for a classification knowledge based system. The originality of VODKA is to identify these new modified objects, the variants, from the way that the existing knowledge based system fails in applying some rules with low certainty degree. In this method, we try to classify the current new evolving object identified according to its attributes and their corresponding values. According to the analysis of the collected inference logs, one of the three recommendations (including adding a new attribute-value of an attribute, modifying the data type of an attribute, or adding a new attribute) will be suggested to help experts observe and characterize the new confirmed variants. VODKA requires E-EMCUD (extended embedded meaning capturing and uncertainty deciding). EMCUD is a knowledge acquisition system which relies on the repertory grids technique to manage object/attribute-values tables and to produce inferences rules from these tables. The E-EMCUD we used here is a new version of EMCUD to update existing tables by adding new objects or new attributes and to adapt the original embedded rules. Here, a computer worm detection prototype is implemented to evaluate the effectiveness of VODKA. The experimental results show that new worm variants could be discovered from inference logs to customize the corresponding detection rules for computer worms. Moreover, VODKA can be applied to the e-learning area to learn the variant learning behaviors of students and to reconstruct the teaching materials in improving the performance of e-learners.

View all citing articles on Scopus

View full text

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

Abstract

International Journal of Man-Machine Studies

Cognitive Science

Probabilistic reasoning and certainty factors

A progress report on the training of probability assessors

The application of artificial intelligence to a maintenance and diagnostic information system (MDIS)

Boeing blackboard system

Personal construct theory and the transfer of human expertise

Rapid acquisition and combination of knowledge from multiple experts in the same domain

Expert systems for design problem-solving using design refinement with plan selection and redesign

Generic tasks in knowledge-based reasoning: the “Right,” level of abstraction for knowledge acquisition

CSRL: a language for classificatory problem solving and uncertainty handling

AI Magazine

Generic tasks in knowledge-based reasoning: high-level building blocks for expert system design

IEEE Expert

In defense of probability

Heuristic classification

MOLE: a tenacious knowledge-acquisition tool

An overview of knowledge acquisition and transfer

International Journal of Man-Machine Studies

New directions in the analysis and interactive elicitation of personal construct systems

Induction of inference rules for expert systems

Fuzzy Sets and Systems

The Dempster-Shafer theory of evidence

Design for acquisition principles of knowledge system design to facilitate knowledge acquisition

International Journal of Man-Machine Studies

Propagating uncertainty by logic sampling in Bayes' networks

The change of personal constructs from the viewpoint of a theory of implications

Influence diagrams

MORE: an intelligent knowledge acquisition tool