Lateral land movement prediction from GNSS position time series in a machine learning aided algorithm

We investigate the accuracy of conventional machine learning aided algorithms for the prediction of lateral land movement in an area using the precise position time series of permanent GNSS stations. The machine learning algorithms that are used are tantamount to the ones used in [1], except for the radial basis functions, i.e. multilayer perceptron, Bayesian neural network, Gaussian processes, k-nearest neighbor, generalized regression neural network, classification and regression trees, and support vector regression. A comparative analysis is presented in which the accuracy level of the mentioned machine learning methods is checked against each other. It is shown that the most accurate method for both of the components of the time series is the Gaussian processes, achieving up to 9.5 centimeters in accuracy.