Title:	THE DESIGN AND DEVELOPMENT OF A MEDICAL KNOWLEDGE MINING SYSTEM
Author(s):	Nittaya Kerdprasop, Kittisak Kerdprasop
ISBN:	978-972-8939-16-8
Editors:	Mário Macedo
Year:	2010
Edition:	Single
Keywords:	Medical knowledge mining system, Intelligent medical informatics, Logic-based knowledge mining.
Type:	Full Paper
First Page:	127
Last Page:	134
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	Healthcare organizations regularly generate huge amount of data in electronic form and store in heterogeneous databases. These data are a valuable resource for automatic discovering of useful knowledge, known as knowledge mining or data mining, to support high-level decisions. Knowledge mining is the process of deriving new and useful knowledge from vast volumes of data and information stored as background knowledge. Derived knowledge may be patterns of data represented in summarized form, relationships among data represented as rules, or representatives of data subgroups represented by mean values. During the past decades there has been an increasing interest in devising database and machine learning technologies to automatically induce knowledge from biomedicine, clinical and health data using imperative and object-oriented programming styles. In this paper, we propose the design and implementation of a knowledge mining system as a repertoire of tools based on the logical framework. Logic programming is a declarative high-level programming scheme suitable for the development of knowledge intensive tasks. Our system includes three major knowledge induction components: data classification, association mining, and data clustering. The tree-based classification, frequent pattern mining, and k-means clustering are implemented with higher-order predicates using Prolog programming language. We also demonstrate knowledge deployment aspects through the trigger creation for the active databases and the automatic generation of a knowledge base for the medical decision support system.
	Go Back