Dr Young-Ok (Joanna) Cha

Research

Natural Language Processing, Large Language Models, Mateial Science, Database building, Data Analysis

Interests

My research interests lie at the intersection of AI-driven materials discovery, machine learning, and electromagnetic systems, with particular emphasis on metamaterials, metasurface and antenna technologies, and additive manufacturing for electromagnetics. I am especially motivated by developing mathematical algorithms, simulation tools, and data-centric AI methodologies to address new, challenging, and interdisciplinary problems in electromagnetic materials and devices.

A central theme of my current research is the application of natural language processing (NLP), large language models (LLMs), and machine learning to extract, structure, and analyse knowledge embedded in scientific literature for accelerated materials discovery. Scientific publications represent a vast yet underutilised data resource; however, their heterogeneous formats which are texts, graphs, and tables pose significant challenges for conventional data-mining techniques. To address this, I have developed an AI-enabled framework that automatically extracts experimental data from the literature and constructs structured databases for predictive modelling. This framework has been successfully applied to extract temperature-dependent resistance ratio data for analysing dopant effectiveness in phase-change materials such as GST and VO₂, enabling downstream machine-learning models for new dopant discovery.

I have extensive research experience across both academia and industry. Prior to my doctoral studies, I worked at Samsung, where I advanced software-defined integration of LCD and touch sensors, focusing on CMOS-based image sensors and signal-processing algorithms for extracting meaningful information from sensing data. This work resulted in 11 international patents and provided me with a strong theoretical and practical foundation in electronic circuits, PCB design, and control systems. I also collaborated closely with global industry partners including Motorola, Sharp, and Hitachi, gaining valuable insight into industrial research culture and collaborative workflows—experience directly relevant to partnerships with organisations such as QinetiQ, Thales UK, and Huawei Technologies.

During my PhD, I led and contributed to multiple NLP- and machine-learning-based research projects, including keyword prediction in metamaterial science, automated summarisation for reviews on organ-specific wireless bioelectric devices, future antenna technology prediction, and automatic data extraction from scientific texts, graphs, and tables. These projects involved constructing large-scale scientific databases (over 159,000 abstracts), applying word embeddings and clustering in high-dimensional spaces, and developing encoder–decoder LSTM models with attention mechanisms to predict research trends and technological evolution in the form of Gartner-style hype cycles.

Through this research journey, I have developed strong expertise in NLP, LLMs, machine learning, and big-data technologies, alongside a solid understanding of metamaterial science and wireless communication systems. I am enthusiastic about pursuing interdisciplinary research that bridges AI, materials science, and electromagnetics to accelerate scientific discovery and technological innovation.

Publications

2023

2022

Research Group

News