Design of a Method to Integrate Knowledge Discovery Techniques with Prior Domain Knowledge for Better Decision Support ABSTRACT
BibTeX
@MISC{Kunene_designof,
author = {K. Niki Kunene and H. Roland Weistroffer},
title = {Design of a Method to Integrate Knowledge Discovery Techniques with Prior Domain Knowledge for Better Decision Support ABSTRACT},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
The current data mining methodology uses knowledge discovery techniques to predict the likelihood of specific events using information residing in databases, or determines specific clusters or classes from some data or text. The knowledge discovered is assumed to be interpretable and useful for decision making purposes; however data mining experts are required to make the interpretation in the context of limited understanding of the business environment, its objectives and constraints, managerial decision preferences and options. This paper describes the design of a method for integrating domain specific decision preferences, options, objectives and constraints into the knowledge discovery process, extending the use of data mining technology to include direct support for decision making by managers. This is accomplished by integrating data mining and knowledge discovery techniques iteratively with rule sets, decision maker specified objective functions, and user qualitative preferences.
Keyphrases
integrate knowledge discovery technique better decision support abstract prior domain knowledge determines specific cluster data mining mining expert objective function decision making discovery technique managerial decision preference information residing business environment specific event current data mining methodology decision maker domain specific decision preference qualitative preference knowledge discovery technique decision making purpose direct support rule set limited understanding knowledge discovery process
