Interpretable AI-driven multidimensional chemical fingerprints for geographical authentication of Euryales Semen

Abstract Euryales Semen (ES, Euryale ferox Salisb.) is a valuable aquatic food in Asia. Its quality and price depend on its geographical origins. To ensure the authenticity of ES, a tracing strategy using stable isotopes, elements, and starch composition with interpretable algorithms was successfully developed. Results indicated that ES from different regions exhibited different chemical fingerprinting profiles. Tree-based intelligent algorithms were introduced for classification, and light gradient boosting machine (LightGBM) achieved the highest accuracy of 97.67%. The SHapley Additive exPlanation (SHAP) interpreted the LightGBM output for feature impact. Notably, the top 10 significant variables, encompassing Na, V, Ba, Sb, Cu, Ti, Mn, %N, amylose, and ratio of amylose to amylopectin (SHAP value >1.0), were selected as the key factors. Moreover, environmental factors were found to be significantly related to these key variables (p < 0.05). Overall, this study offers an effective strategy for the geographical origin traceability of ES or other aquatic crops.

Abstract licence: CC BY-NC-ND

40 more references, whose wording the paper’s licence keeps private.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.