Abstract
We describe a large-scale automatic knowledge acquisition tool, the Quasi-Optimizer (QO) system. It is a domain-independent program that can obtain, verify, fuse and optimize human expertise. The QO system is capable of generating computer models (descriptive theories) of human decision making strategies in two ways. In the passive mode of observation, the system does not or cannot interfere with the environment and records the characteristic features of the situations and the corresponding strategy responses to them. In the active mode of observation, the system designs a sequence of environments (“experiments”) for the decision making strategy to respond to. The design of the experiments can be fixed in advance or can follow a dynamically evolving pattern that minimizes the total number of experiments needed for a userspecified level of precision. A module of QO can ascertain whether a non-static strategy is in fact learning, that is converging to an asymptotic form, to which the program can then extrapolate. Another module can assign a quality measure (credit) to the different strategy components identified by it, on the basis of their short- or long-term benefits in attaining certain goals. The QO system can select the best components of several strategies and combine them in a Super Strategy. The inconsistencies, incompletenesses and redundancies inherent in such a Super Strategy are eliminated and a Quasi-Optimum Strategy is generated. This strategy is better, in the statistical sense, than any one participating in the “training set”. The Quasi-Optimum Strategy, therefore, corresponds to a normative theory within the limitations of the available information. We also describe two rather different areas of application — a possible approach to the automation of air traffic controllers' training and evaluation, and the automatic verification and validation of discrete-event simulation models.
Preview
Unable to display preview. Download preview PDF.
References
Findler, N. V.: An overview of the Quasi-Optimizer system (Large-Scale Systems, Theory and Applications, 5, pp. 123–130; 1983).
Findler, N. V.: On a computer-based theory of strategies (Kybernetes, 12, pp. 89–97; 1983).
Findler, N. V.: On the automatic generation of descriptive and normative theories (Invited Paper; Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, pp. 596–601. Bombay and New Delhi, India; 1983).
Findler, N. V.: Some Artificial Intelligence contributions to air traffic control (Proceedings of the Fourth Jerusalem Conference on Information Technology, pp. 470–475. Jerusalem, Israel; 1984).
Findler, N. V., M. S. Belofsky and T. W. Bickmore: On some issues concerning optimization and decision trees (Proceedings of the International Conference on Mathematical Modelling in Science and Technology, pp. 191–197. Zurich, Switzerland; 1983).
Findler, N. V., T. W. Bickmore and R. F. Cromp: A generalpurpose man-machine environment with special reference to air traffic control (International Journal for Man-Machine Studies, 23, pp. 587–603; 1985).
Findler, N. V., T. Bickmore, L. Ihrig and W.-W. Tsang: On the heuristic optimization of a certain class of decision trees (Information Sciences; 53, pp. 89–100, 1990).
Findler, N. V. and R. F. Cromp: A heuristic approach to optimum experimental design (Computational Statistics and Data Analysis, 2, pp. 167–178; 1984).
Findler, N. V. and J. van Leeuwen: The complexity of decision trees, the Quasi-Optimizer, and the power of heuristic rules (Information and Control, 40, pp. 1–19; 1979).
Findler, N. V. and J. P. Martins: On automating computer model construction — The second step toward a Quasi-Optimizer system (Journal of Information and Optimization Sciences, 2, pp. 119–136; 1981).
Findler, N. V. and N. Mazur: On the verification and validation of discrete-event simulation models (Transactions of the Society for Computer Simulation International, 6, pp. 153–172, 1989).
Findler, N. V., N. Mazur and B. McCall: A note on computing the asymptotic form of a limited sequence of decision trees (Information Sciences, 28, pp. 207–231; 1983).
Findler, N. V. and B. B. McCall: A conceptual framework and a heuristic program for the credit-assignment problem (IEEE Transactions on Systems, Man and Cybernetics, SMC-14, pp. 750–754; 1984).
Findler, N. V., W.-W. Tsang and L. Ihrig: A pruning algorithm for finding all optimal decision graphs (Proceedings of the International Conference on Optimization Techniques and Applications, pp. 470–479. Singapore; 1987).
Findler, N. V.: Contributions to a Computer-Based Theory of Strategies (Chapters 2 and 3. Springer-Verlag: New York, Berlin, Heidelberg; 1990).
Minsky, M.: Steps toward artificial intelligence (In E. A. Feigenbaum and J. Feldman (Eds.): Computers and Thought, pp. 406–450. McGraw-Hill: New York; 1963).
Neumann, J. von and O. Morgenstern: Theory of Games and Economic Behavior (Third ed. Princeton University Press: Princeton, NJ; 1953).
Simon, H. A.: Models of Man (Wiley: New York, NY; 1957).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1992 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Findler, N.V. (1992). An automatic knowledge acquisition tool. In: Mřrík, V., Štěpánková, O., Trappl, R. (eds) Advanced Topics in Artificial Intelligence. Lecture Notes in Computer Science, vol 617. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-55681-8_44
Download citation
DOI: https://doi.org/10.1007/3-540-55681-8_44
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-55681-7
Online ISBN: 978-3-540-47271-1
eBook Packages: Springer Book Archive