See a map of all the places I've given a talk!
I have presented my research at international conferences, workshops, and invited talks. Below are my conference presentations, invited talks, and other speaking engagements. Click on any talk for more details.
Conference presentation, The 10th Asia-Pacific Conference on Wind Engineering (APCWE10), Seoul, South Korea
Presented our research on using Retrieval Augmented Generation (RAG) for assisting engineers in using wind load design codes. The system combines large language models with retrieval mechanisms to provide accurate and context-aware assistance.
Journal publication presentation, Journal of Nondestructive Evaluation, Online
Presented our research on machine learning-assisted automated impact-echo testing for concrete structures. The study investigated the feasibility of ML models for automatic classification of impact-echo test results, achieving reliable detection of shallow delamination and other defect types.
Journal publication presentation, ASCE Journal of Structural Engineering, Online
Presented our novel approach to structural health monitoring using multiclass Deep SVDD. The method achieves 88.11% accuracy in defect detection using only normal condition data, significantly outperforming traditional methods (58.68% accuracy).
Journal publication presentation, ACI Structural Journal, Online
Presented our research on lightweight machine learning models for concrete crack image classification. The study addresses the challenge of balancing accuracy with computational efficiency for on-site applications.
Conference proceedings talk, The 2024 World Congress on Advances in Civil, Environmental, and Materials Research, Seoul, South Korea
This talk presents our research on using multi-sphere deep SVDD for system identification of building structures. The method demonstrates improved performance in detecting structural anomalies and damage patterns compared to traditional approaches.
Conference proceedings talk, The 2024 Structures Congress, Seoul, Korea
This talk presents our research on representation learning for concrete crack image classification. We demonstrate how learned representations can improve classification accuracy and robustness in real-world conditions.
Conference presentation, 8th European-African Conference on Wind Engineering (EACWE2022), Warsaw, Poland
Presented our research on clustering wind pressure taps using dynamic time warping. The method enables efficient sensor placement and wind pressure analysis for high-rise buildings.
Conference presentation, The 2021 World Congress on Advances in Structural Engineering and Mechanics, Seoul, South Korea
Presented our research on predicting wind pressure coefficients on high-rise building facades using LSTM RNN models. The method enables sensor reduction while maintaining prediction accuracy.