ISSN print edition: 0366-6352 ISSN electronic edition: 1336-9075 Registr. No.: MK SR 9/7 Published monthly

Advances in the applications of machine learning for cosmetic formulation development

Wen Jiang, Lingyun Yao, Qingran Meng, and Qianjie Zhang

School of Perfume and Aroma Technology, Shanghai Institute of Technology, Shanghai, People’s Republic of China

E-mail: jw@sit.edu.cn

Abstract:

Machine learning (ML), a core branch of AI, enhances efficiency, accuracy, and systematization in cosmetic formulation development through advanced data processing and analytical capabilities. This paper follows a framework covering database comparisons, ML applications in raw material screening, ingredient matching, production quality control, and personalized cosmetics, supported by analysis of techniques, challenges, and emerging technologies. ML enables virtual screening of functional ingredients, predicts their mechanisms of action, and facilitates multi-factor optimization considering safety, efficacy, and stability. In production, ML systems integrate machine vision and molecular fingerprinting for real-time monitoring and quality defect detection. For personalization, ML-driven skin analysis and consumer preference mining support tailored skincare solutions. Key challenges include data source reliability, appropriate model selection, privacy protection, regulatory compliance, and market acceptance. The future advancements should focus on improving data quality, enhancing model interpretability, strengthening privacy mechanisms, and promoting cross-disciplinary collaboration to fully realize ML’s potential in cosmetic innovation.

Full paper is available at www.springerlink.com.

DOI: 10.1007/s11696-026-04696-7

Chemical Papers 80 (5) 4635–4670 (2026)