고용 빅데이터에서 결과 변수의 계층 불균형 문제를 해결하기 위한 조건부 표 형식의 생성적 적대적 네트워크(GAN)의 응용 Application of Conditional Tabular Generative Adversarial Networks (GAN) for Addressing Class Imbalance in Nationwide Employment Big Data
한국기계기술학회지
(韓國機械技術學會誌)
제26권 제5호
(2024.10)
pp.911-926
pp.911-926
한국기계기술학회
(Korean Society of Mechanical Technology)
초록
This study investigates using Conditional Tabular Generative Adversarial Networks (CT-GAN) to generate synthetic data for turnover prediction in large employment datasets. The effectiveness of CT-GAN is compared with Adaptive Synthetic Sampling (ADASYN), Synthetic Minority Over-sampling Technique (SMOTE), and Random Oversampling (ROS) using Logistic Regression (LR), Linear Discriminant Analysis (LDA), Random Forest (RF), and Extreme Learning Machines (ELM), evaluated with AUC and F1-scores. Results show that GAN-based techniques, especially CT-GAN, outperform traditional methods in addressing data imbalance, highlighting the need for advanced oversampling methods to improve classification accuracy in imbalanced datasets.
키워드
목차
Abstract
1. Introduction
2. Related works
3. Materials and Methods
3.1. Imbalance Ratio (IR)
3.2. Random Oversampling (ROS)
3.3. Synthetic Minority Over-SamplingTechnique (SMOTE)
3.4. B-SMOTE
3.5. Adaptive Synthetic Sampling (ADASYN)
3.6. Conditional GAN (CGAN)
3.7. Conditional Tabular GAN (CT-GAN)
3.8. Modeling
3.9. Data source
3.10. Experimental design
3.11. Performance Evaluation Methods andMetrics
4. Results
5. Discussion
6. Conclusions
References
1. Introduction
2. Related works
3. Materials and Methods
3.1. Imbalance Ratio (IR)
3.2. Random Oversampling (ROS)
3.3. Synthetic Minority Over-SamplingTechnique (SMOTE)
3.4. B-SMOTE
3.5. Adaptive Synthetic Sampling (ADASYN)
3.6. Conditional GAN (CGAN)
3.7. Conditional Tabular GAN (CT-GAN)
3.8. Modeling
3.9. Data source
3.10. Experimental design
3.11. Performance Evaluation Methods andMetrics
4. Results
5. Discussion
6. Conclusions
References
저자