Kernel-based non-paxametric models have been applied widely over recent years. However, the associated computational complexity imposes limitations on the applicability of those methods to problems with large data-sets. In this paper we develop a filtering approach based on a Gaussian process regression model. The idea is to generate a smalldimensional set of filtered data that keeps a high proportion of the information contained in the original large dataset. Model learning and prediction are based on the filtered data, thereby decreasing the computational burden dramatically.