Least Squares Support Vector Machine Regression Based on Sparse Samples and Mixture Kernel Learning

Abstract

Least squares support vector machine (LSSVM) is a machine learning algorithm based on statistical theory. Its
advantages include robustness and calculation simplicity, and it has good performance in the data processing
of small samples. The LSSVM model lacks sparsity and is unable to handle large-scale data problem, this article
proposes an LSSVM method based on mixture kernel learning and sparse samples. This algorithm reduces the
initial training set to a sub-dataset using a sparse selection strategy. It converts the single kernel function in the
LSSVM model into a mixed kernel function and optimizes its parameters. The reduced sub-dataset is used for
training LSSVM. Finally, a group of datasets in the UCI Machine Learning Repository were used to verify the
effectiveness of the proposed algorithm, which is applied to real-world power load data to achieve better fitting
and improve the prediction accuracy.