Multipath is one of the most challenging factors to model and/or characterize in the GNSS observation error budget. For the case of ground stations, code phase static multipath is typically the largest contribution of local observation errors. Current approaches for multipath characterization include the analysis of code-minus-carrier (CMC) observables and the exploitation of multipath repeatability. This contribution presents an alternative strategy for multipath detection and characterization based on unsupervised and self-supervised machine learning methods. The proposed strategy makes use of observations in the Receiver Independent Exchange Format (RINEX), typically generated by GNSS receivers in ground stations, for model training and testing, without requiring the availability of labelled data. To assess the performance of the proposed strategy (data-based), a comparison with a model-based methodology for multipath error prediction using a digital twin model is carried out. Results from a test case using data from a monitoring station of the International GNSS Service (IGS) show a consistency between the two approaches. The proposed methodology is applicable for a similar characterization in any GNSS ground station.