Automated discovery of causal relationships in managerial problem domains /

Bibliographic Details
Main Author:	Billman, Bethany Hindman
Other Authors:	Baysinger, Barry D. (degree committee member.), Olson, David L. (degree committee member.), Paradice, David B. (degree committee member.), Tretter, Marietta J. (degree committee member.)
Format:	Thesis Book
Language:	English
Published:	1989.
Subjects:	Problem solving > Data processing. Decision making > Data processing. Management information systems. Major business analysis. Academic theses
Online Access:	ProQuest, Abstract Link to OAKTrust copy

Description
Abstract:	The research investigated the extent to which the computer itself can propose hypotheses of causal relationships among variables in a business problem domain. The goals of the research were: (1) to develop a methodology for automating three theories for positing hypotheses of causality among data elements in organizational databases, (2) to determine the extent to which these automated discovery methods could be used for development of a knowledge base in a managerial problem domain, and (3) to evaluate the performances of the three automated discovery methods compared with that of human decision makers. This research aimed to improve the managerial problem formulation process and construction of cognitive maps - activities which are inherently biased and restricted by perceptions of decision makers, and by volatility and uncertainty in managerial environments. The results indicated that the three automated discovery methods outperformed experienced decision makers in finding causal relationships in a simulated managerial problem domain.
Item Description:	Typescript (photocopy). Vita. "Major subject: Business Analysis."
Physical Description:	xii, 160 leaves : illustrations ; 29 cm
Bibliography:	Includes bibliographical references.