| dc.creator | Grinblat, Guillermo | |
| dc.creator | Granitto, Pablo M. | |
| dc.creator | Ceccatto, Alejandro | |
| dc.date | 2011-03-11T13:46:59Z | |
| dc.date | 2011-03-11T13:46:59Z | |
| dc.date | 2008 | |
| dc.date | 2011-03-11T13:46:59Z | |
| dc.date | 2011-03-11T13:46:59Z | |
| dc.date | 2008 | |
| dc.date.accessioned | 2019-05-17T20:03:13Z | |
| dc.date.available | 2019-05-17T20:03:13Z | |
| dc.identifier | 1137-3601 | |
| dc.identifier | http://hdl.handle.net/2133/1718 | |
| dc.identifier | http://hdl.handle.net/2133/1718 | |
| dc.identifier.uri | http://repositorioslatinoamericanos.uchile.cl/handle/2250/2676104 | |
| dc.description | In this work we propose an adaptive classification method able both to learn and to follow the temporal
evolution of a drifting concept. With that purpose we introduce a modified SVM classifier, created using
multiple hyperplanes valid only at small temporal intervals (windows). In contrast to other strategies
proposed in the literature, our method learns all hyperplanes in a global way, minimizing a cost function
that evaluates the error committed by this family of local classifiers plus a measure associated to the VC
dimension of the family. We also show how the idea of slowly changing classifiers can be applied to non-linear
stationary concepts with results similar to those obtained with normal SVMs using gaussian kernels. | |
| dc.format | application/pdf | |
| dc.language | en_US | |
| dc.publisher | Asociación Española de Inteligencia Artificial | |
| dc.relation | http://www.aepia.org/ | |
| dc.rights | © AEPIA | |
| dc.rights | Open access | |
| dc.subject | Adaptive methods | |
| dc.subject | Support Vector Machine | |
| dc.subject | Drifting concepts | |
| dc.title | Time–Adaptive Support Vector Machines | |
| dc.type | Artículos de revistas | |
