본문 바로가기
KERD
논문KCI 등재

사례 선택 기법을 활용한 앙상블 모형의 성능 개선

Improving an Ensemble Model Using Instance Selection Method

민성환
한국산업경영시스템학회 (KOREAN SOCIETY OF INDUSTRIAL AND SYSTEMS ENGINEERING)|2016.03|한국산업경영시스템학회지 (Journal of Korean Society of Industrial and Systems Engineering)|제39권 제1호|pp. 105-115

0

조회

0

다운로드

2

인용

0

참고문헌

키워드

Genetic AlgorithmsBaggingBankruptcy PredictionRandom Subspace
주제분류산업공학

초록

Ensemble classification involves combining individually trained classifiers to yield more accurate prediction, compared with individual models. Ensemble techniques are very useful for improving the generalization ability of classifiers. The random subspace ensemble technique is a simple but effective method for constructing ensemble classifiers; it involves randomly drawing some of the features from each classifier in the ensemble. The instance selection technique involves selecting critical instances while deleting and removing irrelevant and noisy instances from the original dataset. The instance selection and random subspace methods are both well known in the field of data mining and have proven to be very effective in many applications. However, few studies have focused on integrating the instance selection and random subspace methods. Therefore, this study proposed a new hybrid ensemble model that integrates instance selection and random subspace techniques using genetic algorithms (GAs) to improve the performance of a random subspace ensemble model. GAs are used to select optimal (or near optimal) instances, which are used as input data for the random subspace ensemble model. The proposed model was applied to both Kaggle credit data and corporate credit data, and the results were compared with those of other models to investigate performance in terms of classification accuracy, levels of diversity, and average classification rates of base classifiers in the ensemble. The experimental results demonstrated that the proposed model outperformed other models including the single model, the instance selection model, and the original random subspace ensemble model.

Abstract

Ensemble classification involves combining individually trained classifiers to yield more accurate prediction, compared with individual models. Ensemble techniques are very useful for improving the generalization ability of classifiers. The random subspace ensemble technique is a simple but effective method for constructing ensemble classifiers; it involves randomly drawing some of the features from each classifier in the ensemble. The instance selection technique involves selecting critical instances while deleting and removing irrelevant and noisy instances from the original dataset. The instance selection and random subspace methods are both well known in the field of data mining and have proven to be very effective in many applications. However, few studies have focused on integrating the instance selection and random subspace methods. Therefore, this study proposed a new hybrid ensemble model that integrates instance selection and random subspace techniques using genetic algorithms (GAs) to improve the performance of a random subspace ensemble model. GAs are used to select optimal (or near optimal) instances, which are used as input data for the random subspace ensemble model. The proposed model was applied to both Kaggle credit data and corporate credit data, and the results were compared with those of other models to investigate performance in terms of classification accuracy, levels of diversity, and average classification rates of base classifiers in the ensemble. The experimental results demonstrated that the proposed model outperformed other models including the single model, the instance selection model, and the original random subspace ensemble model.

참고문헌 (35)

  1. Abellan, J., "Improving Experimental Studies about Ensembles of Classifiers for Bankruptcy Prediction and Credit Scoring", Expert Systems with Applications, 41(8), pp.3825~3830 (2014)
  2. Altman, E. L., "Financial ratios, discriminant analysis and the prediction of corporate bankruptcy", The Journal of Finance, 23(4), pp.589~609 (1968)
  3. Beaver, W., "Financial ratios as predictors of failure, empirical research in accounting : Selected studied", Journal of Accounting Research, 4(3), pp.71~111 (1966)
  4. Bian, S., "On diversity and accuracy of homogeneous and heterogeneous ensembles", International Journal of Hybrid Intelligent Systems, 4(2), pp.103~128 (2007)
  5. Breiman, L., "Bagging predictors", Machine Learning, 24(2), pp.123~140 (1996)
  6. Bryant, S. M., "A case-based reasoning approach to bankruptcy prediction modeling", International Journal of Intelligent Systems in Accounting, Finance and Management, 6(3), pp.195~214 (1997)
  7. Derrac, J., "Enhancing evolutionary instance selection algorithms by means of fuzzy rough set based feature selection", Information Sciences, 186(1), pp.73~92 (2012)
  8. Dietterich, T. G., "Machine-learning research : Four current directions", AI Magazine, 18(4), pp.97~136 (1997)
  9. Freund, Y., "Experiments with a new boosting algorithm", Proceedings of the 13th, International Conference on Machine learning, pp.148~156 (1996)
  10. Garcia, V., "On the use of data filtering techniques for credit risk prediction with instance-based models", Expert Systems with Applications, 39(18), pp.13267~13276 (2012)
  11. Goldberg, D.E., "Genetic algorithms in search, optimization and machine learning" (1989), Addison-Wesley
  12. Hart, P. E., "The condensed nearest neighbor rule", IEEE Transactions on Information Theory, 14, pp.515~516 (1968)
  13. Ho, T. K., "The random subspace method for constructing decision forests", IEEE Transactions on Pattern Analysis and Machine Intelligence, 20(8), pp.832~844 (1998)
  14. Hung, C., "A Selective Ensemble Based on Expected Probabilities for Bankruptcy Prediction", Expert Systems with Applications, 36(3), pp.5297~5303 (2009)
  15. 김경재, "재무예측을 위한 Support Vector Machine의 최적화", 지능정보연구, 17(4), pp.241~254 (2011), 한국지능정보시스템학회
  16. Kim, M., "Ensemble with neural networks for bankruptcy prediction", Expert System with Applications, 37(4), pp.3373~3379 (2010)
  17. Kim, M., "Geometric Mean Based Boosting Algorithm with over-Sampling to Resolve Data Imbalance Problem for Bankruptcy Prediction", Expert Systems with Applications, 42(3), pp.1074~1082 (2015)
  18. Kuncheva, L. I., "Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy", Machine Learning, 51(2), pp.181~207 (2003)
  19. Li, H., "The random subspace binary logit (RSBL) model for bankruptcy prediction", Knowledge-Based Systems, 24(8), pp.1380~1388 (2011)
  20. Louzada, F., "Poly-bagging predictors for classification modelling for credit scoring", Expert Systems with Applications, 38(10), pp.2717~12720 (2011)
  21. Marques, A. I., "Exploring the Behaviour of Base Classifiers in Credit Scoring Ensembles", Expert Systems with Applications, 39(11), pp.10244~10250 (2012)
  22. Messier, W., "Inducing rules for expert system development : an example using default and bankruptcy data", Management Science, 34(12), pp.1403~1415 (1998)
  23. Meyer, P. A., "Prediction of bank failures", The Journal of Finance, 25, pp.853~868 (1970)
  24. Min, S. -H., "Hybrid genetic algorithms and support vector machines for bankruptcy prediction", Expert Systems with Applications, 31(3), pp.652~660 (2006)
  25. Nanni, L., "An Experimental Comparison of Ensemble of Classifiers for Bankruptcy Prediction and Credit Scoring", Expert Systems with Applications, 36(2), pp.3028~3033 (2009)
  26. Ohlson, J., "Financial ratios and the probabilistic prediction of bankruptcy", Journal of Accounting Research, 18, pp.109~131 (1980)
  27. 박경종, "실수 코딩 유전자 알고리즘을 이용한 생산 시스템의 시뮬레이션 최적화", 산업경영시스템학회지, 28(3), pp.149~155 (2005), 한국산업경영시스템학회
  28. Tam, K., "Managerial applications of neural networks : the case of bank failure predictions", Management Science, 38(7), pp.926~947 (1992)
  29. Tsai, C., "Using Neural Network Ensembles for Bankruptcy Prediction and Credit Scoring", Expert Systems with Applications, 34(4), pp.2639~2649 (2008)
  30. Wang, G., "A hybrid ensemble approach for enterprise credit risk assessment based on Support Vector Machine", Expert Systems with Applications, 39(5), pp.5325~5331 (2009)
  32. 유재욱, "유전자 알고리듬을 이용한 소프트웨어 제품라인의 출시 계획 수립", 산업경영시스템학회지, 35(4), pp.142~148 (2012), 한국산업경영시스템학회[DOI]
  33. 엄창선, "유전자 알고리즘을 이용한 경제적 LAN 설계", 산업경영시스템학회지, 28(2), pp.101~108 (2005), 한국산업경영시스템학회
  34. Yum, J. K., "A Study of D-Optimal Design in Nonlinear Model Using the Genetic Algorithm", Journal of the Korean Society for Quality Management, 28(2), pp.135~146 (2000)
  35. Zhang, G., "Artificial neural networks in bankruptcy prediction :general framework and cross-validation analysis", European Journal of Operational Research, 116, pp.16~32 (1999)

관련 논문AI추천

분류기 앙상블 선택을 위한 혼합 유전 알고리즘김영원, 오일석한국정보처리학회 · 정보처리학회 논문지2007 · 제14권 제6호데이터 불균형 해소를 위한 유전알고리즘 기반 최적의 오버샘플링 비율김용혁, 신승수, 조휘연한국융합학회 · 한국융합학회논문지2021 · 제12권 제1호앙상블을 이용한 기계학습 기법의 설계: 보험가입자의 심사평가 예측을 통한 실험적 검증이제동, 최정문한국IT정책경영학회 · 한국IT정책경영학회 논문지2021 · 제13권 제6호더블 앙상블 기법을 이용한 기술사업성평가 기반의 중소기업 부실 예측 연구유동희, 이상훈신용보증기금 · 중소기업금융연구2022 · 제42권 제2호혼합물 실험계획법을 활용한 앙상블 예측 모델권영석, 이광현, 이동주한국산업경영시스템학회 · 한국산업경영시스템학회지2024 · 제47권 제4호신용평가를 위한 데이터마이닝 분류모형의 통합모형에 관한 연구김갑식한국정보처리학회 · 정보처리학회논문지D2005 · 제12권 제2호

이 간행물의 다른 논문 (18건)

특허경영활동이 기업 경영성과에 미치는 영향에 관한 연구 : 국내 의료기기 제조 기업을 중심으로윤장혁, 김용현, 정병기pp. 1-8베이지안넷 기반의 프로젝트 일정리스크 평가성홍석, 박철순pp. 9-16프로젝트 예비비 편성 방법에 관한 연구권혁천, 강창욱pp. 17-24AREA 활용 전력수요 단기 예측권세혁, 오현승pp. 25-30Live-Virtual 시뮬레이션 환경차이에 따른 조종사 스트레스 유발요인 분석홍영석, 김성호, 설현주, 김진주, 지철규pp. 31-36
원문 보기