GIST Repository
College of Engineering
Department of Environment and Energy Engineering
3. Theses(Master)
Predicting Rapid Intensification of Tropical Cyclones in the Western North Pacific Using TabNet Classifier: Performance Improvement through Data Augmentation and Loss Function Optimization
- Abstract
- In this study, the TabNet model was employed to effectively predict Rapid Intensification (RI) cases of Tropical Cyclones (TCs) in the Northwestern Pacific region. By leveraging the model’s interpretability, key features influencing predictions were analyzed, revealing Sea Surface Temperature (SST) as the most significant factor, followed by Temperature and Specific Humidity at 850 hPa. These findings align with existing research, underscoring TabNet’s ability to accurately identify variables critical to ocean-atmosphere interactions. To address the severe class imbalance between RI and non-RI cases, the Synthetic Minority Oversampling Technique (SMOTE) was utilized with ratios of 4:1, 3:1, 2:1, and 1:1, and three loss functions—Cross Entropy, Balanced Cross Entropy, and Focal Loss—were combined to design twelve experimental setups. The combination of Balanced Cross Entropy and 2:1 SMOTE achieved the highest Critical Success Index (CSI) of 0.780, effectively balancing class imbalance and maximizing prediction performance. The optimized TabNet model demonstrated approximately 46.15% higher Probability of Detection (POD) and 42.31% higher CSI compared to the Hurricane Weather Research and Forecasting (HWRF) model, highlighting TabNet’s superior performance in predicting minority classes. Yearly performance evaluation revealed significant performance degradation in 2020, attributed to the scarcity of RI cases rather than atypical meteorological conditions. This emphasizes the critical role of data quantity and diversity in enhancing RI prediction accuracy. Additionally, the high interpretability of the TabNet model facilitates quantitative evaluation of influential variables, enhancing the understanding and reliability of prediction results. In conclusion, this research demonstrates that TabNet, combined with data augmentation and optimized loss functions, significantly improves RI prediction performance, surpassing traditional Numerical Weather Prediction (NWP) models. Future work should focus on further addressing data imbalance and enhancing model robustness under diverse environmental conditions to develop more reliable weather prediction systems. MS/EN 20231076
- Issued Date
- 2025
- Type
- Thesis
- Alternative Author(s)
- Sanghyeok An
- Department
- 대학원 환경에너지공학부
- Advisor
- Yoon, Jin-Ho
- Table Of Contents
- ABSTRACT 1
CONTENTS 2
LIST OF FIGURES 4
LIST OF TABLES 5
1. Introduction 6
2. Literature Review 8
2.1. Advances in TC RI case prediction research 8
2.2. TC RI Prediction Using Machine Learning 9
3. Data & Methods 11
3.1. Data 11
3.1.1. Best Track Data 11
3.1.2. ERA5 Reanalysis Data 11
3.2. Calculated Variables 13
3.3. TabNet Classifier 16
3.3.1. Feature Transformer 16
3.3.2. Sparse Attentive Transformer 17
3.3.3. Decision Prediction Layer 17
3.4. SMOTE Data Augmentation 19
3.5. Loss Function 20
3.5.1. Binary Cross Entropy 20
3.5.2. Balanced Binary Cross Entropy 20
3.5.3. Focal Loss 21
3.6. Validation Metric 22
3
3.6.1. Probability of Detection 23
3.6.2. False Alarm Ratio 23
3.6.3. Critical Success Index 23
4. Results 24
4.1. Experimental design and model selection 24
4.2. Comparison of experimental results 25
4.3. Performance comparison with NWP model 27
4.4. Performance comparison with other machine learning model 28
4.5. Feature Importance Analysis 30
4.6. Performance Analysis by Year 31
5. Discussion 35
6. Conclusion 37
References 39
ACKNOWLEDGEMENT 42
CURRICULUM VITAE 43
4
- Degree
- Master
- Appears in Collections:
- Department of Environment and Energy Engineering > 3. Theses(Master)
- 파일 목록
-
-
관련 파일이 존재하지 않습니다.
-
Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.