Research on Least Squares Support Vector Machine Combinatorial Optimization Algorithm

摘要

LS-SVM (least squares support vector machine) has been widely used in engineering practice. However, the solving of LS-SVM still remains difficult under the condition of large sample. Based on algorithm of combinatorial optimization, this paper put forward the combinatorial optimization least squares support vector machine algorithm. On several different data aggregation of dimensions, the numerical value experiment and comparison are carried out on traditional LS-SVM algorithm, COLS-SVM algorithm and its improvement algorithm. The numerical value test has shown that COLS-SVM algorithm and its improvement algorithm are effective and have certain advantages on time and regression accuracy, compared with traditional LS-SVM algorithm.