Risk-Aware Imitation Learning with Environmental Context for AIS
Imitation-learning anomaly detection for AIS is improved by adding environmental data (wind, waves), so the model can tell apart risky vessel behavior from safe weather-driven detours.
LLM Agent for Maritime Data Analysis
This project develops a Hybrid Prompt Agent that enables natural-language analysis of maritime AIS data by combining query classification with dynamic prompting.
Development of a robust and generalizable foundation model for bioelectrical signals
Foundation model development tailored to the unique characteristics of bioelectrical signal acquisition
Physics-Informed AI for Robust Ship Fuel Consumption Prediction
Development of a Ship Fuel Consumption Prediction Model Considering Environmental Factors
Fault Detection via Domain-Knowledge-Based Training Data Refinement
A domain knowledge–based data refinement methodology for detecting defective products that cannot be filtered out in the LQC process.
Developing data-driven user engagement metrics for contents(functions) in Updatable home appliances
Developing a new metric for measuring customer satisfaction based on user activity data
Deep Learning Approach for Behavior of Piston of Linear Compressor
Deep learning–based sensorless control of linear compressor pistons with over 90% performance improvement.
Personalized Dose Determination for Patients with Thyroid Hormone Disorders
Optimal and Personalized Dose Determination for Patients with Thyroid Hormone Disorders Using Deep Learning-Based Survival Analysis
Domain Knowledge-Informed Functional Outlier Detection for LQC
An ST-based method using failure pattern knowledge to detect tiny anomalies in manufacturing time-series data.
Active Thyroid-Associated Orbitopathy Detection on Frontal Eye Photographs
Developing a deep learning–based AI system for early monitoring and diagnosis of thyroid eye disease to enable timely treatment before irreversible damage.
Real Time Changing-State Quasar Detection
Real-Time Detection of Changing-State Quasars using a Mixture Density Network
Prediction of Traffic Congestion Propagation
Modeling and quantifying time-lagged accident-induced congestion using causal inference and bootstrap uncertainty analysis.
CNN based Gas Mixture Classification
A CNN-based multi-channel time-series analysis method is proposed for accurate classification of gas mixtures using electronic nose sensor data.
Maritime Anomaly Detection
A statistical approach for route planning and anomaly detection to enhance maritime situational awareness.