오차 패턴 모델링을 이용한 Hybrid 데이터 마이닝 기법 = A Hybrid Data Mining Technique Using Error Pattern Modeling

다국어 초록 (Multilingual Abstract)

This paper presents a new hybrid data mining technique using error pattern modeling to improve classification accuracy when the data type of a target variable is binary. The proposed method increases prediction accuracy by combining two different supervised learning methods. That is, the algorithm extracts a subset of training cases that are predicted inconsistently by both methods, and models error patterns from the cases. Based on the error pattern model, the predictions of two different methods are merged to generate final prediction. The proposed method has been tested using practical 10 data sets. The analysis results show that the performance of proposed method is superior to the existing methods such as artificial neural networks and decision tree induction.

참고문헌 (Reference)

1 "입력자료 판별에 의한 데이터 마이닝 성능개선" 293-303, 2000

2 "이동통신고객 분류를 위한 의사결정나무 와 신경망 결합 알고리즘 연구" 139-155, 2003

3 "연관규칙과 퍼지 인공신경망에 기반한 하이브리드 데이터 마이닝 메커니즘에 대한 연구" 884-888, 2003

4 "앙상블 학습알고리즘의 일반화 성능비교：OLA, Bagging, Boosting" (97) : 226-228, 2000

5 "모형평가와 앙상블을 이용한 데이터 마이닝에 관한 연구" 293-306, 1998

6 "데이터 마이닝을 위한 경쟁학습모델과 BP알고리즘을 결합한 하이브리드 신경망" 1-16, 2002

7 Schapire,R, "The Strength of Weak Learnerbility" 5 (5): 197-227, 1990

8 "The Improvement of Response Modeling： Combining Rule-induction and Case-based Reasoning" 18 (18): 307-313, 2000

9 "The Hybrid of Association Rule Algorithms and Genetic Algorithm for Tree Induction：An Example of Predicting the Student Course Performance" 25 (25): 51-62, 2003

10 "Predicting the Exchange Traded Fund DIA with a Combination of Genetic Algorithm and Neural Networks" 27 (27): 417-425, 2004

11 Kuncheva,L.I.C. Bezdek, "On Combining Multiple Classifiers by Fuzzy Templates" 193-197, 1998

12 "Neural Networks Ensembles" 12 (12): 993-1001, 1990

13 Li,R, "Mining Classification Rules Using Rough Sets and Neural Networks" 157 (157): 439-448, 2004

14 "Hybrid Decision Tree/Genetic Algorithm Method for Data Mining" 163 (163): 13-35, 2004

15 "Generalized Additive Multi-mixture Model for Data Mining" 38 (38): 487-500, 2002

16 "Experiments with a New Boosting Algorithm" 148-156, 1996

17 Indurkhya,N, "Estimating Performance Gains for Voted Decision Trees" 2 (2): 303-310, 1998

18 "Error Reduction through Learning Multiple Descriptions" 24 (24): 105-112, 1996

19 "Ensembling Neural Networks：Many Could Be Better Than All" 137 (137): 239-263, 2002

20 "Customer List Segmentation Using the Combined Response Model" 17 (17): 89-97, 1999

21 "Combining Classification Algorithms" Departamento de Ciecia de Computadores Faculdade de Ciecias da Universidade do Porto 1999.

22 "Clementine Ver. 8 User’s Guide" SPSS Inc 2003.

23 "Clementine Stream Prototypes:Part 2" 1-7, 2004

24 "Boosting the Margin：A New Explanation for the Effectiveness of Voting Methods" 322-330, 1998

25 Quinlan, R, "Bagging, Boosting and C4.5" AAAI Press 1996.

26 "Bagging Predictors" 24 (24): 123-140, 1996

27 Lin,F.Y, "A Data Mining Approach to the Prediction of Corporate Failure" 14 (14): 189-195, 2001

28 "A Comparison of Three Strategies to Rule Induction from Data with Numerical Attributes" 82 (82): 1-9, 2003