Research
Research Overview
My research combines materials science, machine learning, and data analytics to develop practical systems for bio-detection, wearable electronics, and interpretable modeling.
Focus Areas
- Nanomaterials for bio-detection and wearable systems.
- Applied machine learning for material-property prediction and process optimization.
- Smart textile systems including flexible antennas and conductive fabrics.
- Data-driven experimental design for robust and repeatable research outcomes.
Selected Research Projects
Gold Nanoparticle Size and Surface Prediction
Objective: Predict size and surface properties of gold nanoparticles from growth factors.
Methods: Random Forest and Gradient Boosting on experimental datasets.
Outcome: Strong predictive performance (R2 0.997, RMSE 0.348) and interpretable feature importance.
Baby Name Trend Forecasting with Machine Learning
Objective: Forecast baby-name popularity using historical trends and cultural signals.
Methods: Logistic regression and feature clustering for time-aware classification.
Outcome: High-accuracy modeling and multi-year forecast insights for demographic analysis.
Research Experience
Biophysics Lab, North Carolina State University (2017-2018)
Conducted DNA bio-detection research with upconversion nanoparticles and analyzed time-series spectroscopy datasets to study molecular motion.Smart Textile Research, North Carolina State University (2018-2022)
Led collaborations on textile-based antennas and nanomaterial inks, combining spectroscopy, optimization, and machine learning for wearable electronics.Donghua University (2013-2016)
Synthesized Ag nanoparticles and improved bio-detection performance through spray-coating workflows and regression-based optimization.
Recent Publications
Textile-Integrated Liquid Metal Electrodes for Electrophysiological Monitoring (2022, Advanced Healthcare Materials)
Project page | DOIDesign of Quasi-Endfire Spoof Surface Plasmon Polariton Leaky-Wave Textile Wearable Antennas (2022, IEEE ACCESS)
Project page | DOIInterconnected Cathode-Electrolyte Double-Layer Enabling Continuous Li-Ion Conduction Throughout Solid-State Li-S Battery (2022, Energy Storage Materials)
Project page | DOI
Methods and Tools
- Data science: Python, Pandas, NumPy, scikit-learn, MATLAB.
- Modeling: Random Forest, Gradient Boosting, Logistic Regression, clustering.
- Characterization: FTIR, Raman, UV-Vis, fluorescence spectroscopy.
- Statistics: ANOVA, regression analysis, time-series analysis.