Yoshiyuki Kobayashi, Ph.D.

Lecturer, Laboratory of Medical Data Science, Faculty of Pharmacy and Pharmaceutical Sciences, Hoshi University, Tokyo, Japan

Research

My field is informatics in the pharmaceutical sciences. I collect and analyze the data that governments and companies rely on when they make decisions about chemicals in medicines, pesticides and food. Current topics:

• Machine-learning QSAR/QSPR models for the environmental fate of pesticides and other chemicals
• Exposure assessment of food allergens, and international comparison of regulations for health foods and biostimulants
• Large language model-assisted text mining (LATeM) of research trends
• Drug development, orphan drug lag and digital health, with members of our laboratory

Before moving to academia, I spent more than ten years in the pharmaceutical, agrochemical and consulting industries, including pesticide registration for overseas markets.

Appointments

• 2026– Lecturer, Hoshi University
• 2025– Visiting Associate Professor, Research Organization of Open Innovation and Collaboration, Ritsumeikan University
• 2025– Visiting Researcher, Graduate School of Medicine, Kyoto University
• 2025– Collaborative Researcher, School of Medicine, Keio University
• 2025–2026 Assistant Professor, Hoshi University

Degrees

• Ph.D. in Engineering, University of Tsukuba (2026)
• Ph.D. in Pharmaceutical Sciences, Hoshi University (2021)
• Doctor of Business Administration, University of Tsukuba (2021)

Grant

JSPS KAKENHI Grant-in-Aid for Early-Career Scientists (25K18320), FY2025–FY2026, principal investigator. The project compares biostimulant regulations across countries and examines how they affect innovation and market development.

Selected publications

1. Kobayashi Y, Inoue T, Kageyama I, et al. Comprehensive Analysis of Food Allergy Research Trends via Large Language Model–Assisted Text Mining (1980–2024). Allergy. 2025;81(4):1281–1284. doi:10.1111/all.70169
2. Kobayashi Y, Inoue T, Suzuki Y, et al. Exposure estimation of the food allergens milk and wheat protein in Japan through Bayesian methods: implications for food labeling. Journal of Food Composition and Analysis. 2025;148(1):108078. doi:10.1016/j.jfca.2025.108078
3. Kobayashi Y, Uchida T, Kageyama I, et al. Uncovering new psychoactive substances research trends using large language model-assisted text mining (LATeM). Journal of Hazardous Materials Advances. 2025;18:100700. doi:10.1016/j.hazadv.2025.100700
4. Kobayashi Y, Uchida T, Inoue T, et al. A Comprehensive Analysis of the per- and poly-fluoroalkyl substances (PFAS) research landscape through AI-assisted text mining. Journal of Hazardous Materials Letters. 2024;5. doi:10.1016/j.hazl.2024.100121
5. Kobayashi Y, Yoshida K. Development of QSAR models for prediction of fish bioconcentration factors using physicochemical properties and molecular descriptors with machine learning algorithms. Ecological Informatics. 2021;63:101285. doi:10.1016/j.ecoinf.2021.101285
6. Kobayashi Y, Uchida T, Yoshida K. Prediction of Soil Adsorption Coefficient in Pesticides Using Physicochemical Properties and Molecular Descriptors by Machine Learning Models. Environmental Toxicology and Chemistry. 2020;39(7):1451–1459. doi:10.1002/etc.4724

Full list: Publications (32 papers) / ORCID / Google Scholar / researchmap

Contact

Email: contact [at] yoshiyukikobayashi.com

A note on the design: the Japanese pages imitate Japanese personal homepages from the early 2000s, the middle of the Heisei era. The retro look is on purpose.