In this paper, we provide a theoretical framework for feature selection in tree kernel spaces based on gradient-vector components of kernel-based machines. We show that a huge number of features can be discarded without a significant decrease in accuracy. Our selection algorithm is as accurate as and much more efficient than those proposed in previous work. Comparative experiments on three interesting and very diverse classification tasks, i.e. Question Classification, Relation Extraction and Semantic Role Labeling, support our theoretical findings and demonstrate the algorithm performance. 1
Scheda prodotto non validato
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte di FBK.
Titolo: | On Reverse Feature Engineering of Syntactic Tree Kernels |
Autori: | |
Data di pubblicazione: | 2010 |
Abstract: | In this paper, we provide a theoretical framework for feature selection in tree kernel spaces based on gradient-vector components of kernel-based machines. We show that a huge number of features can be discarded without a significant decrease in accuracy. Our selection algorithm is as accurate as and much more efficient than those proposed in previous work. Comparative experiments on three interesting and very diverse classification tasks, i.e. Question Classification, Relation Extraction and Semantic Role Labeling, support our theoretical findings and demonstrate the algorithm performance. 1 |
Handle: | http://hdl.handle.net/11582/10508 |
Appare nelle tipologie: | 4.1 Contributo in Atti di convegno |