论文部分内容阅读
受小波理论与再生核Hilbert空间理论的启发,提出了一种新的小波再生核。该小波再生核由不同分辨率的小波基函数生成,并且是一种容许的支持向量核。应用该小波再生核,构造了用于函数学习的最小二乘支持向量回归模型。这种回归模型融合了支持向量机与小波的优点。仿真例子说明了该方法的可行性与有效性。
Inspired by wavelet theory and Hilbert space theory, a new wavelet kernel is proposed. The wavelet regeneration kernel is generated by wavelet bases of different resolutions and is an allowable kernel of support vector. By using this wavelet kernel, a least square support vector regression model for function learning is constructed. This regression model combines the advantages of SVM and wavelet. The simulation example shows the feasibility and effectiveness of this method.