Application Research of the Robust LS-SVM Regression Model in Forecasting Patent Application Quantities

摘要

A forecasting system of patent application quantities is studied by means of applying the support vector machine (SVM) which uses cross-validation algorithms to select preferences. The result of data emulation shows the proposed method has higher forecasting precision and stronger generalization ability than BP neural network and RBF neural network, and it is feasible and effective in forecasting patent application quantities.