In this article, we will describe how the kernel approach can be easily implemented for simple and typical knowledge discovery problems, within the context of machine learning. Since the core of this paradigm relies on the so called kernel trick, we will mainly focus on how this fundamental tool can be effectively used, in the design and the application of an inference procedures. In fact, the kernel approach has not only offered to the learning machine community the opportunity of working both with nonlinear predictive models and with different heterogeneous structures, but it has also given a new way to re-design old standard procedures, in order to get more powerful and relative robust models.
Zoppis, I., Mauri, G., Dondi, R. (2019). Kernel Machines: Applications. In S. Ranganathan, M. Gribskov, K. Nakai, C. Schönbach (a cura di), Encyclopedia of Bioinformatics and Computational Biology : ABC of Bioinformatics. Vol.1: Methods (pp. 511-518). Cambridge : Elsevier [10.1016/B978-0-12-809633-8.20343-9].
Kernel Machines: Applications
Zoppis, Italo;Mauri, Giancarlo;Dondi, Riccardo
2019
Abstract
In this article, we will describe how the kernel approach can be easily implemented for simple and typical knowledge discovery problems, within the context of machine learning. Since the core of this paradigm relies on the so called kernel trick, we will mainly focus on how this fundamental tool can be effectively used, in the design and the application of an inference procedures. In fact, the kernel approach has not only offered to the learning machine community the opportunity of working both with nonlinear predictive models and with different heterogeneous structures, but it has also given a new way to re-design old standard procedures, in order to get more powerful and relative robust models.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
