This paper introduces a procedure that transforms multiple evaluation metrics into a single aggregated score, providing a comprehensive and interpretable summary of machine learning performance. The approach is demonstrated on a set of metrics obtained from various anomaly detection algorithms based primarily on Isolation Forest. Seven relevant performance metrics are aggregated using diverse techniques, including the arithmetic mean, weighted mean, Choquet integral, the OWA operator, and several Smooth OWA variants based on different interpolation Newton-Cotes quadratures. For methods requiring them, two distinct sets of weights are used. The results show that, particuraly in anomaly detection tasks where individual metrics may lead to inconsistent evaluations, the aggregated score effectively reflects metric preferences and enables quick identification of the best-performing algorithm for a given dataset.