Journal of Intelligence and Information Systems (지능정보연구)

Korea Intelligent Information System Society (한국지능정보시스템학회)
권호 표지

Data Mining using Instance Selection in Artificial Neural Networks for Bankruptcy Prediction

기업부도예측을 위한 인공신경망 모형에서의 사례선택기법에 의한 데이터 마이닝

  • Published : 2004.06.01
Download PDF
⟨ Previous Next ⟩

Abstract

Corporate financial distress and bankruptcy prediction is one of the major application areas of artificial neural networks (ANNs) in finance and management. ANNs have showed high prediction performance in this area, but sometimes are confronted with inconsistent and unpredictable performance for noisy data. In addition, it may not be possible to train ANN or the training task cannot be effectively carried out without data reduction when the amount of data is so large because training the large data set needs much processing time and additional costs of collecting data. Instance selection is one of popular methods for dimensionality reduction and is directly related to data reduction. Although some researchers have addressed the need for instance selection in instance-based learning algorithms, there is little research on instance selection for ANN. This study proposes a genetic algorithm (GA) approach to instance selection in ANN for bankruptcy prediction. In this study, we use ANN supported by the GA to optimize the connection weights between layers and select relevant instances. It is expected that the globally evolved weights mitigate the well-known limitations of gradient descent algorithm of backpropagation algorithm. In addition, genetically selected instances will shorten the learning time and enhance prediction performance. This study will compare the proposed model with other major data mining techniques. Experimental results show that the GA approach is a promising method for instance selection in ANN.

기업부도예측은 재무와 경영의사결정문제에서의 주된 인공신경망 응용분야라 할 수 있다. 일반적으로 인공신경망은 이 분야에서 매우 좋은 성과를 보이는 것으로 알려져 있지만 종종 잡음이 심한 데이터에 대해서는 일관성 있고 예측가능한 성과를 보이지 못하는 경우가 있다. 특히 학습용 자료가 매우 많아서 학습시간과 자료수집비용이 과대한 경우에는 적절한 자료의 축소가 되지 않고는 인공신경망을 학습시키는 것이 불가능한 경우도 있다. 사례선택기법은 자료의 차원을 축약시켜 주며 직접적으로 자료를 축소시켜 주는 방법이다. 사례기반 학습기법에서는 이미 몇 연구가 사례선택기법의 필요성을 주장한 바 있으나 인공신경망 모형에서 사례선택기법의 필요성을 주장한 연구는 거의 없다. 본 연구에서는 기업부도예측을 위한 인공신경망 모형에서 유전자 알고리즘을 이용한 사례선택기법을 제안한다. 본 연구에서 유전자 알고리즘은 다층 인공신경망에서의 계층별 연결강도를 최적화하고, 동시에 학습에 적합한 사례를 선택한다. 유전자 알고리즘에 의해 결정된 계층별 연결강도는 역전파오류 학습기법에서 종종 발생하는 국부 최적해에 수렴하는 현상을 최소화해 줄 것으로 기대되고, 선택된 학습용 사례는 학습시간의 단축과 예측성과를 향상시켜 줄 것으로 기대된다. 본 연구에서는 제안한 모형과 주요 데이터 마이닝 기법들의 성과를 비교 연구한다. 실험결과, 제안된 방법이 인공신경망에서의 사례선택기법으로 유용한 것으로 나타났다.

Keywords

References

  1. Machine learning:Neural networks,genetic algorithms,and fuzzy systems Adeli,H.;S.Hung
  2. The Journal of Finance Financial ratios,discrominant analysis and the prediction of corporate bankruptcy Altman,E.I.
  3. Journal of Banking and Finance v.18 Corporate distress diagnosis:Comparisons using linear discriminant analysis and neural networks Altman,E.I.;G.Macro;F.Varetto
  4. Genetic algorithms and investment strategies Bauer,R.J.
  5. Artficial Intelligence v.97 no.1-2 Selection of relevant features and examples in machine learning Blum,A.;P.Langley
  6. Expert Systems with Applications v.9 no.4 Effectiveness of neural network types for prediction of business failure Boritz,J.E.;D.B.Kennedy
  7. Intelligent Systems in Accounting,Finance and Management v.4 Predicting corporate failure using a neural network approach Boritz,J.E.;D.B.Kennedy;A.de Miranda e Albuquerque
  8. International Journal of Intelligent Systems in Accounting,Finance and Management v.9 Artificial neural networks in accounting and finance:Modeling issues Coakley,J.R.;C.E.Brown
  9. Business research methods Cooper,D.R.;C.W.Emory
  10. Nearest Neighbor(NN) Norms:NN Pattern Classification Techniques Dasarathy,B.V.
  11. Genetic algorithms and financial applications, in Trading on the edge Davis,L.;G.J.Deboeck(ed.)
  12. Journal of Accounting Research v.Spring A discriminant analysis of predictors of business failure Deakin,E.
  13. Comparing backpropagation with a genetic algorithm for neural network training,Omega v.27 no.6 Gupta,J.N.D.;R.S.Sexton
  14. Statistical methods for business and economics,Reading Harnett,D.R.;A.K.Soni
  15. Computational Intelligence v.15 no.3 Time series prediction with genetic-algorithm designed neural networks:An empirical comparison with modern statistical models Hansen,J.V.;J.B.McDonald;R.D.Nelson
  16. European Journal of Operational Research v.126 no.1 A framework for the description of evolutionary algorithms Hertz,A.;D.Kobler
  17. Expert Systems with Applications v.11 no.4 Integration of case-based forecasting,neural network and discriminant analysis for bankruptcy prediction Jo,H.;I.Han
  18. Expert Systems with Applications v.13 no.2 Bankruptcy prediction using case-based reasoning,neural network and discriminant analysis Jo,H.;I.Han;H.Lee
  19. Expert Systems with Appications v.19 no.2 Genetic algorithms approach to feature discretization in artificial neural networks for the prediction of stock price index Kim,K.;I.Han
  20. Neurocomputing v.21 no.1-3 Predicting bankruptcies with the self-organizing map Kiviluoto,K.
  21. Pattern Recognition Letters v.16 no.8 Editing for the k-nearest neighbors rule by a genetic algorithm Kuncheva,L.I.
  22. Decision Support Systems v.18 Hybrid neural network models for bankruptcy predictions Lee,K.C.;I.Han;Y.Kwon
  23. IEEE Intelligent Systems and Their Applications v.13 no.2 Feature transformation and subset selection Liu,H.;H.Motoda
  24. Data reduction via instance selection,in Instance selection and construction for data mining Liu,H.(ed.);H.Motoda(ed.)
  25. Neural Computing & Applications v.1 Self-organizing neural networks for the analysis and representation of data:Some financial cases Martin del Brio,B.;C.Serrano Cinca
  26. Knowledge-Based Systems v.13 no.2-3 Automating case selection in the construction of a case library McSherry,D.
  27. Expert Systems v.12 no.1 Knowledge acquisition and adaptation:a genetic approach Odetayo,M.O.
  28. Proceedings of the IEEE International Conference on Neural Networks A neural network model for bankruptcy prediction Odom,M.;R.Sharda
  29. Journal of Accounting Research v.18 no.1 Financial ratios and the probailistic prediction of bankruptcy Ohlson,J.
  30. Using genetic algorithms for training data selection in RBF networks, in Instance selection and construction for data mining Reeves,C.R.;D.R.Bush;H.Liu(ed.);H.Motoda(ed.)
  31. Selection of training sets for neural networks by a genetic algorithm, in Parallel problem-solving from nature-PPSN Ⅴ, Reeves,C.R.;S.J.Taylor;A.E.Eiden(ed.);T.Back(ed.);M.Schoenauer(ed.);H.P.Schwefel(ed.)
  32. Decision Support Systems v.17 Self organizing neural networks for financial diagnosis Serrano Cinca,C.
  33. The European Journal of Finance v.3 no.3 Feedforword neural networks in the classification of financial information Serrano Cinca,C.
  34. Decision Support Systems v.22 no.2 Toward global optimization of neural networks:A comparison of the genetic algorithm and backpropagation Sexton,R.S.;R.E.Dorsey;J.D.Johnson
  35. Proceedings of the $11^th$International Conference on Industrial & Engineering Applications of Artificial Intelligence & Expert Systems Case-base maintenance Smyth,B.
  36. Management Science v.38 no.7 Managerial applications of the neural networks:The case of bank failure predictions Tam,K.Y.;Kiang,M.Y.
  37. Predictive data mining:A practical guide Weiss,S.M.;N.Indurkhya
  38. Machine Learming v.38 Reduction techniques for instance-based learning algorithms Wilson,D.R.;T.R.;Martinez
  39. Decision Support Systems v.11 Bankruptcy prediction using neural networks Wilson,R.L.;R.Sharda
  40. Hybrid neural,genetic,and fuzzy systems,in Trading on the edge Wong,F.;C.Tan,C.;G.J.Deboeck(ed.)
  41. Journal of Business Research v.44 Probabilistic neural networks in bankruptcy prediction Yang,Z.R.;M.B.Platt;H.D.Platt