Parallel algorithm of local support vector regression for large datasets

Abstract

We propose the new parallel algorithm of local support vector regression (local SVR), called kSVR for effectively dealing with large datasets. The learning strategy of kSVR performs the regression task with two main steps. The first one is to partition the training data into k clusters, followed which the second one is to learn the SVR model from each cluster to predict the data locally in the parallel way on multi-core computers. The kSVR algorithm is faster than the standard SVR for the non-linear regression of large datasets while maintaining the high correctness in the prediction. The numerical test results on datasets from UCI repository showed that our proposed kSVR is efficient compared to the standard SVR.