This paper reports on an effort to integrate symbolic and mathematical models to tailor the output of a mathematical model to the particular domain of a decision maker. AESOP combines the Black-Scholes model of stock options pricing with an expert system; the integrated model is designed for use by an options specialist on the American Stock Exchange. The specialist makes a number of adjustments to the output of the mathematical model; the purpose of the symbolic model is to make as many of these modifications as possible automatically. The paper reports on the development and structure of AESOP and presents data on its use.
The ability to learn or adapt is widely recognized as one of the most prominent abilities of any animate or inanimate intelligent system. While considerable progress has been made in the science and technology of machine learning, little of that has been incorporated in traditional knowledge-based systems such as diagnostic or expert systems operating in a managerial environment. In this paper a conceptual model of an adaptive expert system is proposed as an attempt to lay a foundation for building knowledge-based systems that can learn by interacting with the environment. In contrast to existing models for learning (such as for knowledge acquisition and skill refinement) where the issue of noise and uncertainty is usually neglected, our model incorporates a stochastic environment and a learning response behavior which too is stochastic in nature.
Is-a relationships are widely recognized as conveying important information for database design. Although these relationships are implicitly hierarchical in nature, it is not uncommon to find collections of is-a relationships that form nonhierarchical structures. This paper formally defines is-a structures and classifies and interprets them. It illustrates how is-a structures can be used to identify possible database design errors or inefficiencies and to produce a database design that incorporates more of the semantics of an application.
Development of large-scale models often involves—or, certainly could benefit from—linking existing models. This process is termed model integration and involves two related aspects: (1) the coupling of model representations, and (2) the coupling of the processes for evaluating, or executing, instances of these representations. Given this distinction, we overview model integration capabilities in existing executable modeling languages, discuss current theoretical approaches to model integration, and identify the limiting assumptions implicitly made in both cases. In particular, current approaches assume away issues of dynamic variable correspondence and synchronization in composite model execution. We then propose a process-oriented conceptualization and associated constructs that overcome these limiting assumptions. The constructs allow model components to be used as building blocks for more elaborate composite models in ways unforeseen when the components were originally developed. While we do not prove the sufficiency of the constructs over the set of all model types and integration configurations, we present several examples of model integration from various domains to demonstrate the utility of the approach.
This study compares a linear keyword language interface and a restricted natural language interface for data retrieval by a novice user. The comparison focuses on the effect of different data base interfaces on user performance (as measured by query correctness and query writing time) in a query writing task across varying query types and training levels. To accomplish this objective, a laboratory experiment was conducted using a split-plot factorial design using two between-subjects factors and one within-subjects factor. The results indicate that the restricted natural language subjects performed significantly better than the linear keyword language subjects in terms of both query correctness and query writing time.
The research literature to date on electronic meeting systems (EMS) has been dominated by studies of high structure systems. High and low structure are defined here in terms of the preponderance of influence that technology and groups exert on each other. High structure EMS attempt to directly influence group structure and processes through explicit rules and procedures embedded in the software. Low structure EMS do not explicitly build in rules or procedures that govern group interaction. Low structure EMS are hypothesized to increase task focus through the use of shared flexible software tools and shared views of joint work. Results of two experiments, that compare low structure EMS-supported groups to manually-supported groups on equality of participation, degree of task focus, task performance and member satisfaction, for two different tasks, are reported. The experiments found that this low structure EMS had no effect on participation equality or member satisfaction. Contrary to the hypotheses, the EMS was found to decrease task focus. The EMS led to marginally better task performance on a simple evaluative task, and to worse performance on a complex generative task. A revision of the study's theoretical model is developed which takes into account the nature of the task.
Group decision support systems have been advocated as mechanisms for facilitating conflict management in groups. Two noted shortcomings of research on GDSS effects motivate this study: (i) Most researchers have compared the effects of computer-based and manual delivery of structures for supporting group decision making. By treating the GDSS as a “black box,” researchers have neglected attention toward examining the effects of specific capabilities delivered by a GDSS. (ii) Despite the volume of accumulated research, scant attention has been paid to examining GDSS impacts on the group interaction process itself. This research proposes a conceptual view of a GDSS as providing communication and consensus capabilities for supporting the cognitive conflict management process in group decision making. Through a manipulation of the delivery of communication and consensus structures to groups working on a strategic planning task, several exploratory research questions were examined. The GDSS used in this study was the SAMM system. Key results obtained were: (a) despite using the same GDSS structures, groups exhibited a variety of patterns of conflict management processes; (b) the delivery of communication and consensus structures together, as opposed to primarily communication structures, however, did result in higher confrontiveness, or the ability of groups to confront their conflicts and resolve them in positive ways; (c) higher confrontiveness resulted in higher levels of post-meeting consensus; and, (d) computerization of structures enabled groups to confront their conflict and resolve it more positively than when groups were provided with equivalent manual structures. The results demonstrate the value of process-oriented methodological approaches to investigating effects of GDSS designs.
The information systems (IS) development activity in large organizations is a source of increasing cost and concern to management. IS development projects are often over-budget, late, costly to maintain, and not done to the satisfaction of the requesting user. These problems exist, in part, due to the organization of the IS development process, where information systems development is typically assigned by the user (principal) to a systems developer (agent). These two parties do not have perfectly congruent goals, and therefore a contract is developed to specify their relationship. An inability to directly monitor the agent requires the use of performance measures, or metrics, to represent the agent's actions to the principal. The use of multiple measures is necessary given the multi-dimensional nature of successful systems development. In practice such contracts are difficult to develop satisfactorily, due in part to an inability to specify appropriate metrics. This paper develops a principal-agent model that provides a set of decision criteria for the principal to use to develop an incentive compatible contract for the agent. These criteria include the precision and the sensitivity of the performance metric. After presenting the formal model, some current software development metrics are discussed to illustrate how the model can be used to provide a theoretical foundation and a formal vocabulary for performance metric analysis. The model is also used in a positive (descriptive) manner to explain why current practice emphasizes metrics that possess relatively high levels of sensitivity and precision. Finally, some suggestions are made for the improvement of current metrics based upon these criteria.
The optimal amount of information needed in a given decision-making situation lies somewhere along a continuum from “not enough” to “too much”. Ackoff proposed that information systems often hinder the decision-making process by creating information overload. To deal with this problem, he called for systems that could filter and condense data so that only relevant information reached the decision maker. The potential for information overload is especially critical in text-based information. The purpose of this research is to investigate the effects and theoretical limitations of extract condensing as a text processing tool in terms of recipient performance. In the experiment described here, an environment is created in which the effects of text condensing are isolated from the effects of message and individual recipient differences. The data show no difference in reading comprehension performance between the condensed forms and the original document. This indicates that condensed forms can be produced that are equally as informative as the original document. These results suggest that it is possible to apply a relatively simple computer algorithm to text and produce extracts that capture enough of the information contained in the original document so that the recipient can perform as if he or she had read the original. These results also identify a methodology for assessing the effectiveness of text condensing schemes. The research presented here contributes to a small but growing body of work on text-based information systems and, specifically, text condensing.
This paper reports the results of a study that examined the way that professionals used a particular decision aid, spreadsheet software, to address a business-oriented task. The professionals were divided into four groups, depending on whether they were experts or novices in the functional area of business represented by the task and on whether they were experts or novices in the use of spreadsheet software. The physical interactions as well as the thought processes underlying these interactions were captured. Results of the study indicated that a lack of expertise in spreadsheet software usage inhibited the application of functional area knowledge. The behavior and outcomes of the functional area expert/spreadsheet novice group did not match those of the functional area expert/spreadsheet expert group. In fact, they paralleled more closely those of the other two groups that did not possess functional area knowledge. Results of the study also indicated that individuals need to obtain a level of expertise in using a decision aid before they are able to apply their functional area knowledge to the problem at hand.