Intelligent Algorithms and Machine Learning
Our laboratory focuses on building intelligent algorithms that support human decision-making. By integrating optimization theory with statistical learning theory, we aim to design new algorithms that can handle complex real-world problems. Special emphasis is placed on machine learning methods that derive meaningful insights from limited information and on mathematically efficient optimization models with strong theoretical foundations.
Data Science for Management and Marketing
We apply data-driven approaches to enhance managerial decision-making and marketing strategies. Using empirical data such as consumer surveys, sales histories, and economic indicators, we perform predictive and causal analyses to support effective business strategies. In particular, our research employs interpretable machine learning models (e.g., SHAP and ICE) to provide transparent and actionable insights that lead to more convincing and reliable decision-making.
Mathematical Evaluation Methods for Decision Support
Our work also explores quantitative evaluation methods for selecting optimal solutions under multiple criteria. Centered on Data Envelopment Analysis (DEA) and its extended models, we develop frameworks for measuring efficiency, productivity, and performance from multiple perspectives. Furthermore, we focus on constructing robust evaluation indicators that account for uncertainty and variability in real-world data, ensuring reliable assessments in complex decision-making environments.
