Patterns of PCB-138 Bioaccumulation in Small Pelagic Fish from the Eastern Mediterranean Sea Using Explainable Machine Learning Prediction Andreja Stoji´ c, Bosiljka Musta´ c, Gordana Jovanovi´ c, Jasna - Dinovi´ c Stojanovi´ c, Mirjana Periši´ c, Svetlana Staniši´ c, and Snježana Herceg Romani´ c Abstract Fish consumption, especially consumption of oily marine species, is increasing globally due to its recommendation by dieticians. This is due to high polyunsaturated ω-3 and ω-6 (PUFAs) fatty acid content in the tissue of the fish. The health benefits of PUFA ingestion coincide with the risk of intaking haz- ardous lipophilic persistent pollutants including organochlorine pesticides (OCPs) and related polychlorinated biphenyls (PCBs). We examined the impact of 17 fatty acids (FAs) and 36 toxic organic and inorganic contaminants on the behavior patterns of the indicator congener PCB-138 in marine fish using eXtreme Gradient Boost- ing (XGBoost), SHapley Additive exPlanations (SHAP), and SHAP value fuzzy clustering. XGBoost indicated non-linear relationships between PCB-138 and other investigated variables that were explained by SHAP values. The ten obtained fuzzy A. Stoji´ c(B ) · G. Jovanovi´ c · M. Periši´ c Institute of Physics Belgrade, National Institute of the Republic of Serbia, University of Belgrade, Belgrade, Serbia e-mail: andreja@ipb.ac.rs G. Jovanovi´ c e-mail: gordana.vukovic@ipb.ac.rs M. Periši´ c e-mail: mirjana.perisic@ipb.ac.rs B. Musta´ c Department of ecology, agronomy and aquaculture, University of Zadar, Zadar, Croatia e-mail: bmustac@unizd.hr J. - Dinovi´ c Stojanovi´ c Institute of Meat Hygiene and Technology, Ka´ canskog 13, 11 000 Belgrade, Serbia e-mail: jasna.djinovic@inmes.rs S. Staniši´ c Environment and Sustainable Development, Singidunum University, Belgrade, Serbia e-mail: sstanisic@singidunum.ac.rs S. Herceg Romani´ c Institute for Medical Research and Occupational Health, Ksaverska cesta 2, PO Box 291, 10001 Zagreb, Croatia e-mail: sherceg@imi.hr © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 E. Pap (ed.), Artificial Intelligence: Theory and Applications, Studies in Computational Intelligence 973, https://doi.org/10.1007/978-3-030-72711-6_10 175