Behavioral Market Modeling
Modeling consumer preferences, adoption intentions, and demand shifts in emerging technology markets.
#Consumer-DemandMy research examines how emerging technology markets evolve under uncertainty by connecting three layers: technology trajectories, consumer interpretation and adoption, and market-level dynamics that can be explored through data-driven simulation and LLM-based agents. Across mobility, energy, and finance, I use behavioral market modeling, semantic and network-based analysis, and applied data intelligence to turn complex signals into decision-relevant insight.
Modeling consumer preferences, adoption intentions, and demand shifts in emerging technology markets.
#Consumer-DemandUsing persona-based LLM agents to examine decision-making under alternative market scenarios.
#Agent-SimulationAnalyzing how technologies evolve and converge using patent, text, and network-based methods.
#Innovation-SupplyBuilding data-driven analytical systems for mobility, energy, finance, and domain-specific decision support.
#Data-AnalyticsThis manuscript develops a theory-guided, persona-based LLM agent framework for examining heterogeneous consumer decision pathways.
My Master's Thesis on integrating human cognitive reasoning with Large Language Models to simulate and forecast emerging market dynamics.
Ongoing research: Analyzing consumer preferences and psychological thresholds for adopting smart home Demand Response (DR) services using choice experiments.
Presented in Winter Conference of the Journal of the Korean Data Analysis Society. Designing an LLM-based chatbot architecture to extract intent, strictly structure user data into JSON, and automate proactive feedback.
Presented in Winter Conference of the Journal of the Korean Data Analysis Society. An empirical study profiling actual users of a real estate tax solution using RFM variables and K-Means clustering to derive targeted CRM strategies.
This manuscript is currently under review. It develops a semantic-enhanced main path analysis approach for examining technology convergence in smart grids.
Creative Award at the K-Data Science Consortium. Utilizing TRIZ, Neo4j, and LangChain to map technological convergence and simulate R&D ideation through a Knowledge Graph of patents.
Published in the Journal of Korea Society of Industrial Information Systems. An Autoencoder-based approach to detect electrical anomalies in EV batteries, analyzing the effects of charging methods and aging.
Published in the Journal of the Korean Society of Innovation. An empirical study analyzing how MaaS adoption drives demand for shared micromobility in first-last mile trips and its quantitative environmental benefits.
This manuscript is in press. It extends the conference study on MaaS adoption and mobility demand modeling.
Recipient of the Best Paper Award at the 2024 KOTIS Fall Conference. Following expert feedback, this pilot study has been extensively revised and submitted to an international journal.
Excellence Award at the 1st Baekgyeong Hackathon. CATCH! NEWSFIN is an AI agent utilizing EEVE-Korean LLM and RAG to curate financial news and combat financial illiteracy.
Oral presentation archive from the 2023 KSOI Fall Conference. Following the collection of diverse academic feedback, this research has been refined and submitted to a journal.
Published in the Korean Journal of Financial Management. Integrating unstructured cognitive data (news sentiment via BERT) into quantitative financial portfolio optimization.
Excellence Award at Pukyong National University. A data-driven analysis project to optimize urban fire response infrastructure using Sparse PCA and clustering algorithms.