Support Vector Machines with Reject Option for Unequal Costs and Imbalanced Datasets

• 발행기관: 한국자료분석학회
• 저자명: Kanghyun Lee Donghwa Yeo
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.26 No.6, 1689~1705쪽, 전체 17쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2024.12.31

국문 초록

We consider a binary classification problem based on support vector machines (SVM) with a reject option in nonstandard situations (from the perspective of typical algorithmic assumptions), where it is assumed that misclassification costs differ depending on the class label and that sampling biases exist. In particular, we develop a method for using an SVM with a reject option for handling unequal costs and imbalanced datasets by utilizing a Fisher-consistent surrogate loss function that we propose. Furthermore, we substantiate the proposed method by designing an ��₁-penalty SVM with a reject option (denoted as the modified ��₁-SVMⓇ) for unequal costs and imbalanced datasets. An entire solution path algorithm for the modified ��₁-SVMⓇ is also developed. Experimental results and real data analysis indicate that the proposed method demonstrates good classification performance in the presence of unequal costs and imbalanced datasets.

목차

1. Introduction

2. Methodology: A Review

3. SVM with a Reject Option for Unequal Costs and Imbalanced Datasets

4. Numerical Results

5. Conclusion

References

키워드

support vector machine eject option unequal cost imbalanced data entire solution path algorithm

참고문헌

• Journal of Machine Learning Research
• Computational Statistics and Data Analysis
• IEEE Transactions on Information Theory
• The Annals of Statistics
• Proc. the First International Workshop on Pattern Recognition with Support Vector Machines
• The Annals of Statistics
• Open Systems and Information Dynamics
• Journal of the Korean Data Analysis Society
• Journal of the Korean Data Analysis Society
• Pattern Recognition Letters
• Doctoral Dissertation
• Journal of the Korean Data Analysis Society
• Pattern Recognition Letters
• Proc. the 14th Annual Conference on Computational Learning Therory and 5th European Conference on Computational Learning Theory
• Journal of the Royal Statistical Society: Series B
• Proc. the Joint IAPR International Workshops on Advances in Pattern Recognition
• Electronic Journal of Statistics
• Journal of the Korean Data Analysis Society
• Advances in Neural Information Processing Systems