Measurement of structure deformation is one of the two most important elements in assessing the current operating condition of a hydro-technical facility, which is especially important when the object is under constant expansion. This is the case of KGHM’s Zelazny Most tailing dam which is the largest tailings storage facility (TSF) in Europe. The considerable size of the facility entails a very complex monitoring system consisting of numerous inclinometers, piezometers, seismic stations, geodetic benchmarks, etc. Interpretation of data from such an extensive system requires a certain degree of automation. It is not possible to perform a real-time complete data analysis through human resources, despite several teams responsible for supervision and maintenance of the TSF. The detection of anomalous events is one of the objectives of the monitoring process. This problem concerns, among others, the readings of the inclinometers responsible for the measurement of surface displacements, necessary in the assessment of tailing dam stability. The article presents methods of finding anomalies on the inclinometer with the use of machine learning techniques, which significantly simplifies the process of identifying attention-requiring areas. The effectiveness of the algorithms was tested on data samples from various measurement points. The best method will be to build learning-based supervised classifiers in the decision-making process of the TSF stability.