Modelling Local Strontium Isotope Baselines from Human Isotopic Data Using Machine Learning: Implications for Past Mobility Studies - Training Dataset

This dataset is designed for training machine learning models combined with Bayesian statistical approaches. It is part of paper “Modelling Local Strontium Isotope Baselines from Human Isotopic Data Using Machine Learning: Implications for Past Mobility Studies” It contains measurements of stable isotope values together with their original locality-based interpretations, which were used as labels for training classical classification algorithms. The dataset enables the development and evaluation of models that integrate isotopic geochemical signals with probabilistic inference for improved classification and localization tasks. The file contains two worksheets. List 1 (Literature list) provides an overview of sampling locations, including the number of samples per locality and corresponding references to the relevant literature. List 2 (Samples) contains individual samples with their associated isotopic measurements and feature values used directly for model training and evaluation. Values in “localness” means original interpretation. 1 = non-local, 0 = local.