Abstract：The support vector machine is applied to predict the surrounding deformations of tunnel. Support vector machine is a new method of machine learning which is applicable to solve the nonlinear problem with small samples and high dimension. It is difficult to predict the surrounding deformations using the general method. The deformations of surrounding rockmasses are taken as a nonlinear deformation time series，then the deformations are preformed by using technique of time series analysis based on support vector machine. The nonlinear relationship among deformation time series is established，and the future deformations is predicted. The results show that this method is feasible and convenient.
