K-Nearest Neighbor algorithm for selection of new gelatin-carboxymethylcellulose films with TiO₂ nanoparticles and propolis extract with antioxidant and light barrier activity

Abstract

Nanoparticles associated with antioxidant ingredients emerge as a new strategy to improve food packaging properties. In this study, we developed gelatin-carboxymethylcellulose films containing propolis extract and TiO2 nanoparticles (1 % and 5 %), and the K-Nearest Neighbor algorithm was used to select the best packaging composition. Films showed excellent antioxidant activity and those mixed with nanoparticles performed better with light and water barrier properties. The film with propolis and 5 % TiO2 nanoparticles showed higher antioxidant activity and opacity, lower water vapor permeability, solubility, water activity, moisture content, and thickness. These results indicate that the KNN algorithm, an artificial intelligence tool, is promising for classifying and selecting films with properties of interest. The films containing propolis extract and TiO2 nanoparticles have excellent properties that validate their application as food packaging.