The research on artificial intelligence-based Deer Velvet Antler traceability model based on emergent data features

Deer velvet antler(DVA) is highly valued for its nutritional properties, but its complex origins and widespread adulteration in the market have led to significant quality discrepancies. Therefore, this study aims to establish an efficient and accurate method for identifying different species of DVA to enhance classification accuracy and traceability. A total of 120 samples were collected from four species of deer antler velvet: Sika Deer Velvet Antler (SVA), Wapiti Velvet Antler (WVA), Reindeer Velvet Antler (RVA), and Moose Velvet Antler (MVA). Multidimensional features such as color, texture, odor, and composition were extracted using Computer Vision, Ultra-fast Gas Phase Electronic Nose, and High-Performance Liquid Chromatography (HPLC) techniques. Through multivariate statistical analysis (VIP >1, P < 0.05), 162 key discriminative factors were identified. Based on the principle of emergence, a classification model, Whale Optimization Algorithm- Random Forest (WOA-RF), was developed by combining Whale Optimization Algorithm (WOA) and Random Forest (RF) to optimize the classification process. The results demonstrated that the WOA-RF model achieved a 100 % success rate in classifying different DVA species. The proposed intelligent classification algorithm, based on the fusion of multidimensional data, not only enables highly efficient identification of DVA species but also reveals the emergent effects generated by the interaction of multidimensional features combined with optimization algorithms. This approach significantly surpasses the limitations of single techniques and provides crucial technical support and methodological guidance for food species identification and traceability.