The shortage of oil resources is currently a global problem. The use of renewable resources instead of non-renewable ones has become a hot topic of research in the eyes of scientists. In the food industry, there is a lot of interest in bio-based smart active packaging that meets the concept of sustainability and ensures safety. The packaging has antibacterial and antioxidant properties that extend the shelf life of food. Its ability to monitor the freshness of food in real time is also beneficial to consumers’ judgement of food safety. This paper summarises the main raw materials for the preparation of bio-based smart active packaging, including proteins, polysaccharides and composite materials. The current status of the preparation method of bio-based smart active packaging and its application in food preservation is summarised. The future development trend in the field of food packaging is foreseen, so as to provide a reference for the improvement of bio-based smart active packaging materials.

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.