Semi-supervised extreme learning machine with wavelet kernel

Abstract

Extreme learning machine (ELM) not only is an effective classifier in supervised learning, but also can be applied on unsupervised learning and semi-supervised learning. The model structure of semi-supervised machine learning machine (SS-ELM) is same as ELM, the difference between them is the cost function. In this paper, we introduce kernel function to SS-ELM and proposed semi-supervised extreme learning machine with kernel (SS-KELM). Wavelet analysis has the characteristics of multivariate interpolation and sparse change, and the wavelet kernel is a kind of multidimensional wavelet function that can approximate arbitrary nonlinear functions. Therefore, we propose semi-supervised extreme learning machine with wavelet kernel (SS-WKELM) based on the wavelet kernel function and SS-ELM. The experimental results show the feasibility and validity of SS-WKELM in classification.