Deformation of railway infrastructure is a critical factor affecting transportation safety. Continuous automated monitoring of railway infrastructure deformation is essential, with particular emphasis on evaluating the quality of monitoring data. This study, on the basis of data quality evaluation indicators such as GNSS data completeness, multipath error, and cycle slip ratio, nine evaluation indicators are selected in accordance with the characteristics of railway infrastructure deformation monitoring, and a predictive model using ridge regression analysis is proposed to assess data quality by correlating multiple quality indicators with final deformation calculation accuracy, thereby providing an absolute evaluation of the quality of GNSS observation data for railway infrastructure deformation. After determining the ridge coefficient, the observation data from eight railway slope monitoring sites are utilized to analyse the performance of model, which demonstrates a root mean square error of 1.1 mm, indicating both the reliability of the model and feasibility of the evaluation method.