Fuente: Foods - Revista científica (MDPI) Foods, Vol. 14, Pages 4255: Prediction of Esterification and Antioxidant Properties of Food-Derived Fatty Acids and Ascorbic Acid Based on Machine Learning: A Review
Foods doi: 10.3390/foods14244255
Authors:
Xinyu Wang
Jianyi Wang
Xiaoyu Zhang
Tiantong Lan
Jingsheng Liu
Hao Zhang

This study is dedicated to summarizing and performing an in-depth analysis of the antioxidant properties of ascorbic acid fatty acid esters. The esterification reaction mechanism of ascorbic acid with palmitic acid, lauric acid, and oleic acid in food systems was elaborated in detail, and its antioxidant mechanism was discussed in depth. The free radical scavenging mechanism and oxidative inhibition effect of two mainstream determination methods, DPPH and ABTS, were analyzed. Esterification, as a core organic synthesis reaction, is widely used in the production of food antioxidants, pharmaceutical ingredients, chemical polymers, and cosmetic oil-based matrices. At the same time, in view of the wide application of machine learning as a multidisciplinary core technology, this paper selects free radical scavenging rate and esterification yield as characteristic parameters and normalizes the offspring into random forest model training to achieve accurate prediction of antioxidant performance. Finally, in the future, it is necessary to expand the data set, optimize the model structure, explore multi-model fusion to improve the prediction effect, and promote the application of machine learning in the screening design of new antioxidants and the optimization of green synthesis processes to promote the intelligent and sustainable development of food antioxidant research.