If the Kernel trick is used to make non linearly separable data separable by moving them to a high dimensional space, which can helps a lot in the case of classification.
How can this work in the case of regression
Stéphane Breton
Classification and regression tasks may be performed based on the same computational engine (i.e. kernel trick) but from distinct viewpoints. Follow:
- Braga, " Improving the kernel regularized least squares method for small-sample regression ", 2015 - http://www.producao.usp.br/bitstream/handle/BDPI/50885/2686880.pdf;jsessionid=A1EA10995ACA6340331B34306B7E68ED?sequence=1
- [Regression vs Classification] Zhang et al., " Wavelet Support Vector Machine ", 2004 -
Article Wavelet Support Vector Machine
0 votes
0 thanks
Stéphane Breton
Additional references of interest dedicated to SVR:
- Stulp et al., "Many regression algorithms, one unified model - A review", 2015 - https://hal.archives-ouvertes.fr/hal-01162281v2/document
- Smola et al., " A tutorial on support vector regression ", 2004 - https://alex.smola.org/papers/2004/SmoSch04.pdf
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Data Mining
- Classification
More Imene Mansour's questions See All
Similar questions and discussions