Login

Publications  •  Project Statistics

Glossary  •  Schools  •  Disciplines
People Search: 
   
Title/Abstract Search: 

Dissertation Information for James Allen Reggia

NAME:
- James Allen Reggia

DEGREE:
- Ph.D.

DISCIPLINE:
- Computing Science

SCHOOL:
- University of Maryland, College Park (USA) (1981)

ADVISORS:
- None

COMMITTEE MEMBERS:
- None

MPACT Status: Incomplete - Not_Inspected

Title: KNOWLEDGE-BASED DECISION SUPPORT SYSTEMS: DEVELOPMENT THROUGH KMS

Abstract: "This dissertation describes the design and implementation of KMS (Knowledge Management System), an experimental domain-independent software laboratory for developing knowledge-based decision support systems.

KMS achieves two major objectives. First, KMS synthesizes a variety of knowledge representation and processing techniques together within a common conceptual framework (problem-oriented semantic networks). These techniques currently include rule-based deduction (KMS.PS), statistical pattern classification (KMS.BAYES), linear discriminant and other scoring functions (KMS.CALC), and frame-based hypothetico-deductive inference generation (KMS.HT). KMS is ""user-friendly"" in the sense that it is directly usable by an application specialist for constructing decision support systems after a brief training period. This direct usability is made possible by high-level, non-procedural knowledge representation languages and a standardized knowledge acquisition process.

The second major achievement of KMS is a new, domain-independent model of diagnostic reasoning (in KMS.HT). This component of KMS is based on the DESCRIPTION as the basic unit of knowledge organization and emphasizes a clear separation of causal and non-causal associations. A hypothetico-deductive inference mechanism maintains parallel adequate and coherent ""explanations"" for known manifestations. In this fashion, the model is able to handle reasonably difficult diagnostic problems where multiple, simultaneous causative factors are present and context-sensitive scoring is required. The development of this new model of diagnostic reasoning establishes the conceptual framework upon which KMS is based as an appropriate one in which to pursue basic research questions in knowledge engineering.

KMS has already been used to construct several decision support systems and some of these are illustrated in this dissertation. It is currently being using in an educational program on computer-assisted medical decision making, and to construct a prototype ""intelligent textbook of neurology"" that combines both decision making and instructional functions within a single framework. Based on this experience, factors important in the selection of an appropriate representation and inference method when building a DSS are identified and an initial set of guidelines are proposed for this selection process."

MPACT Scores for James Allen Reggia

A = 0
C = 1
A+C = 1
T = 0
G = 0
W = 0
TD = 0
TA = 0
calculated 2008-01-31 06:32:17

Advisors and Advisees Graph