지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술대회자료
- 저자정보
- 발행연도
- 2013.6
- 수록면
- 155 - 162 (8page)
이용수
초록· 키워드
오류제보하기
The prediction model is of course the main factor affecting the performance of a knowledge-based system for bankruptcy prediction. Prediction-modeling studies that began with the building of a single best model have progressed to hybrid and ensemble techniques. In ensemble modeling, which combines the outputs of multiple models, the combination scheme is an important issue affecting prediction accuracy. This paper proposes a confidence-based selection approach as part of an ensemble bankruptcy-prediction scheme that can measure unified confidence, even if ensemble members produce different types of continuous-valued outputs. The present experimental results show that when varying the number of models to combine, according to the creation type of ensemble members, the proposed combination method offers the best performance in the ensemble having the largest number of models, even when compared with the methods most often employed in bankruptcy prediction.
목차
Abstract
1. Introduction
2. Review of related studies
3. Research Methodology
4. Model Development
5. Experiments
6. Conclusions
References
참고문헌 (1)
참고문헌 신청
이 논문의 저자 정보
최근 본 자료
댓글(0)
0
UCI(KEPA) : I410-ECN-0101-2014-000-002751470