Earticle

초록

영어

The process of data mining comprises of seven major steps: (1) data integration, (2) data transformation, (3) data cleaning, (4) data selection, (5) pattern extraction or knowledge mining, (6) pattern evaluation, and (7) knowledge presentation. Steps 1 to 4 are pre-data mining, whereas steps 6 and 7 may be viewed as post-data mining. Therefore, the seven major steps can be grouped into pre-data mining, mining, and post-data mining. We focus our study on the post-data mining processing. Most data mining systems finish their processing at the knowledge presentation step. Our work further the regular post-data mining processing to the step of automatic knowledge deployment. This paper illustrates the knowledge deployment step in which its input is the induced knowledge, in the formalism of classification rules. These rules are evaluated and filtered on the basis of coverage measurement, that is from all the training cases, how many cases are covered by the rule. High coverage rules are transformed into decision rules to be used by the inference engine of the expert system. This post-data mining processing leads to a new design of the next generation rule-based expert system in a medical domain. It is a new idea in that in addition to the set of predefined rules in the knowledge base, the system includes rules that are automatically induced from the database instances. We design the inductive expert system such that the inductive process has been done through the tree-based knowledge discovery technique. Probabilistic decision rules are then transformed from the induced decision tree. The induced, as well as predefined, rules together form a knowledge base for the inductive expert system. Another feature of our system is the inference engine that can be created automatically. The system is intended to support decision making in biomedical informatics. The accuracy of recommendation given by the expert system is evaluated and compared to other three classification systems: decision-tree induction, rule induction, and neural network. The experimental results confirm the high accuracy of our inductive expert system and the automatically created knowledge base.

목차

Abstract
 1. Introduction
 2. Related Work
 3. Probabilistic Knowledge Induction System: Its Design and Methodology
 4. System Implementation and Experimental Results
  4.1. Data Format
  4.2. Prolog Implementation: Knowledge Induction and Probabilistic Rule Generation
  4.3. Prolog Implementation: Automatic Inference Engine and Knowledge Base Creation
  4.4. Performance Evaluation Result on Post-operative Data
  4.5. Performance Evaluation Result on Breast-cancer Recurrence Data
 5. Conclusion
 Acknowledgement
 References

키워드

저자

• Kittisak Kerdprasop [ Data Engineering Research Unit, School of Computer Engineering, Suranaree University of Technology ]
• Nittaya Kerdprasop [ Data Engineering Research Unit, School of Computer Engineering, Suranaree University of Technology ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Bio-Science and Bio-Technology

• 간기

  격월간

• pISSN

  2233-7849

• 수록기간

  2009~2016

• 등재여부

  SCOPUS

• 십진분류

  KDC 505 DDC 605