Research Areas
Home > Research > Research Areas
Embedded Files
The Industrial Data Science Lab (IDSL) at Hanyang University, led by Professor Misuk Kim, focuses on developing industrial data analysis technologies based on Natural Language Processing (NLP) and Large Language Models (LLMs).
Our lab conducts practical research on tasks such as ESG information extraction, Conversational AI systems, Safe-AI, Ontology-based Knowledge Graph and Recommender, aiming to create real-world, industry-applicable NLP solutions by combining cutting-edge LLM techniques with traditional NLP methods.
Autonomous Ontology Construction and Domain-specific Knowledge Agent
Autonomous Ontology Construction and Domain-specific Knowledge Agent
Our lab develops technologies for autonomous ontology construction and domain-specific knowledge agents to enable structured and intelligent use of industrial data. Its research covers data-driven ontology construction from heterogeneous industrial data, document and knowledge graph–based retrieval-augmented generation (RAG), and the development of domain-specific knowledge agents that support efficient knowledge utilization and decision-making in industrial environments.
Conversational AI and LLM Applications
Conversational AI and LLM Applications
Our lab researches the development and optimization of conversational AI systems based on LLMs. Its work includes hierarchical command processing through multi-level prompting, Retrieval-Augmented Generation (RAG)-based question answering systems, hallucination mitigation techniques, and dialogue state tracking for accurate conversation management. The goal is to develop reliable conversational AI solutions that can be practically applied in industrial settings.
ESG and Sustainability Analysis
ESG and Sustainability Analysis
Our lab develops technologies for automatically extracting and classifying ESG-related information from various text sources such as corporate sustainability reports, news articles, and disclosures. Covering the entire process of ESG information handling, from building Korean ESG datasets to developing domain-specific language models, it provides scalable solutions applicable to both large enterprises and small and medium-sized businesses.
Page updated
Google Sites
Report abuse