Title: Barium stars as tracers of s-process nucleosynthesis in AGB stars II. Using machine learning techniques on 169 stars

Abstract: We aim to analyse the abundance pattern of 169 Barium (Ba) stars, using machine learning techniques and the AGB final surface abundances predicted by Fruity and Monash stellar models. We developed machine learning algorithms that use the abundance pattern of Ba stars as input to classify the initial mass and metallicity of its companion star using stellar model predictions. We use two algorithms: the first exploits neural networks to recognise patterns and the second is a nearest-neighbour algorithm, which focuses on finding the AGB model that predicts final surface abundances closest to the observed Ba star values. In the second algorithm we include the error bars and observational uncertainties to find the best fit model. The classification process is based on the abundances of Fe, Rb, Sr, Zr, Ru, Nd, Ce, Sm, and Eu. We selected these elements by systematically removing s-process elements from our AGB model abundance distributions, and identifying those whose removal has the biggest positive effect on the classification. We excluded Nb, Y, Mo, and La. Our final classification combines the output of both algorithms to identify for each Ba star companion an initial mass and metallicity range. With our analysis tools we identify the main properties for 166 of the 169 Ba stars in the stellar sample. The classifications based on both stellar sets of AGB final abundances show similar distributions, with an average initial mass of M = 2.23 MSun and 2.34 MSun and an average [Fe/H] = -0.21 and -0.11, respectively. We investigated why the removal of Nb, Y, Mo, and La improves our classification and identified 43 stars for which the exclusion had the biggest effect. We show that these stars have statistically significant different abundances for these elements compared to the other Ba stars in our sample. We discuss the possible reasons for these differences in the abundance patterns.