Abstract: The deformation monitoring index is an important index for evaluating and monitoring dam safety. Firstly, an improved fast Myriad filtering method is applied to pretreat the deformation monitoring data of the typical dam sections of a RCC gravity dam in the alpine region, and then the maximum entropy method and the cloud model method are used to investigate and analyze the monitoring index of deformation during operation period. Finally, the relationship between the abnormal probability and the contribution rate of weak periphery elements of the cloud model to the qualitative concept is discussed in this study. The analysis results show that the monitoring index of deformation investigated by the maximum entropy method differs from that of the cloud model method; moreover, the maximum measured deformation of the dam during monitoring period is less than the index of the deformation prepared by the two methods. Both of them are calculated by numerical eigenvalues, and the subjective components are little, so the monitoring index of deformation is more credible. The monitoring index of deformation of the weak peripheral elements of the cloud model is close to the monitoring index of deformation under the conditions of 2.15% of abnormal probability. It is considered that the calculation of the contribution rate of the weak periphery elements of the cloud model to the qualitative concept can provide a reference for the abnormal probability determination of the monitoring index of deformation.