Parameters optimization of support vector machine using energy-saving cuckoo search

摘要

Support vector machine (SVM) is a powerful machine learning method for classification and regression problems of small samples and high dimensions. However, in the procedure of classification by SVM, penalty factor c and kernel parameter g have great effect on the performance of classificatioa The parameters selection of SVM has earned widespread respect and some optimization methods have been put forward to deal with it, still, it is not fully solved. Thus the present study proposes a novel meta-heuristic method to optimize parameters of SVM based on energy-saving cuckoo search. In the end, experimental results show that the algorithm seems superior to genetic algorithm and particle swam optimization algorithm and could obtain satisfactory accuracy.