Title page for ETD etd-03042004-105746


Document Type Master's Dissertation
Author De Kock, Erika
Email edk@global.co.za
URN etd-03042004-105746
Document Title Decentralising the codification of rules in a decision support expert knowledge base
Degree MSc Computer Science
Department Computer Science
Supervisor
Advisor Name Title
Prof JM Bishop
Keywords
  • Intelligent Decision Support Systems
  • Knowledge-Based Decision Support Systems
  • Expert Systems
  • Knowledge Management
  • Knowledge Representation
  • Decision Support Generator
  • Decision Support
  • Decision Support Systems
  • Knowledge base
  • Rule-Based Systems
Date 2004-01-30
Availability unrestricted
Abstract
The paradigm of Decision Support Systems (DSS) is to support decision-making, while an Expert Systemís (ES) major objective is to provide expert advice in specialised situations. Knowledge-Based DSS (KB-DSS), also called Intelligent Decision Support Systems (IDSS), integrate traditional DSS with the advances of ES. A KB-DSSí knowledge base usually contains knowledge expressed by an expert and captured by a knowledge engineer. The indirect transfer between the domain expert and the knowledge base through a knowledge engineer may lead to a long and inefficient knowledge acquisition process.

This thesis compares 11 DSS packages in search of a (KB-) DSS generator where domain experts can specify and maintain a Specific Decision Support System (SDSS) to assist users in making decisions. The proposed (KB-) DSS-generator is tested with a university and study-program prototype. Since course and study plan programs change intermittently, the (KB-) DSSí knowledge base enables domain experts to set and maintain their course and study plan rules without the assistance of a knowledge engineer. Criteria are set to govern the (KB-) DSS generator search process. Example knowledge base rules are inspected to determine if domain experts will be able to maintain a set of production rules used in a student registration advice system. By developing a prototype and inspecting knowledge base rules, it was found that domain experts would be able to maintain their knowledge in the decentralised knowledge base, on condition that the objects and attributes used in the rule base were first specified by a builder/programmer.

Files
  Filename       Size       Approximate Download Time (Hours:Minutes:Seconds) 
 
 28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access 
  00Front.pdf 178.67 Kb 00:00:49 00:00:25 00:00:22 00:00:11 < 00:00:01
  01Chapter1.pdf 222.74 Kb 00:01:01 00:00:31 00:00:27 00:00:13 00:00:01
  02Chapter2.pdf 492.17 Kb 00:02:16 00:01:10 00:01:01 00:00:30 00:00:02
  03Chapter3.pdf 200.99 Kb 00:00:55 00:00:28 00:00:25 00:00:12 00:00:01
  04Chapter4.pdf 491.50 Kb 00:02:16 00:01:10 00:01:01 00:00:30 00:00:02
  05Chapter5.pdf 377.64 Kb 00:01:44 00:00:53 00:00:47 00:00:23 00:00:02
  06Chapter6.pdf 241.86 Kb 00:01:07 00:00:34 00:00:30 00:00:15 00:00:01
  07Chapter7.pdf 191.94 Kb 00:00:53 00:00:27 00:00:23 00:00:11 00:00:01
  08Chapter8.pdf 87.17 Kb 00:00:24 00:00:12 00:00:10 00:00:05 < 00:00:01
  09Bibliography.pdf 76.19 Kb 00:00:21 00:00:10 00:00:09 00:00:04 < 00:00:01
  10AppendixA.pdf 136.42 Kb 00:00:37 00:00:19 00:00:17 00:00:08 < 00:00:01

Browse All Available ETDs by ( Author | Department )

If you have more questions or technical problems, please Contact UPeTD.