Classification of transformed anchovy products based on the use of element patterns and decision trees to assess traceability and country of origin labelling

Quadrupole inductively coupled plasma mass spectrometry (Q-ICP-MS) and direct mercury analysis were used to determine the elemental composition of 180 transformed (salt-ripened) anchovies from three different fishing areas before after packaging. To this purpose, four decision trees-based algorithms, corresponding C5.0, classification regression trees (CART), chi-square automatic interaction detection (CHAID), quick unbiased efficient statistical tree (QUEST) applied datasets find most accurate data mining procedure achieve ultimate goal fish origin prediction. Classification rules generated by trained CHAID model optimally identified unlabelled testing bulk (93.9% F-score) using just 6 out 52 elements (As, K, P, Cd, Li, Sr). The finished packaged product was better modelled QUEST algorithm which recognised with F-score 97.7%, considering information carried 5 (B, As, K. Pd). Results obtained suggested that traceability system in fishery sector may be supported simplified machine learning techniques a limited but effective number inorganic predictors origin.