Abstract
Almost all tree kernels proposed in the literature match substructures without taking into account their relative positioning with respect to one another. In this paper, we propose a novel family of kernels which explicitly focus on this type of information. Specifically, after defining a family of tree kernels based on routes between nodes, we present an efficient implementation for a member of this family. Experimental results on four different datasets show that our method is able to reach state of the art performances, obtaining in some cases performances better than computationally more demanding tree kernels.
| Original language | English |
|---|---|
| Title of host publication | ACM International Conference Proceeding Series |
| Volume | 382 |
| DOIs | |
| Publication status | Published - 2009 |
| Externally published | Yes |
| Event | 26th Annual International Conference on Machine Learning, ICML'09 - Montreal, QC Duration: 14 Jun 2009 → 18 Jun 2009 |
Other
| Other | 26th Annual International Conference on Machine Learning, ICML'09 |
|---|---|
| City | Montreal, QC |
| Period | 14/6/09 → 18/6/09 |
ASJC Scopus subject areas
- Human-Computer Interaction
Cite this
Route kernels for trees. / Aiolli, Fabio; Martino, Giovanni; Sperduti, Alessandro.
ACM International Conference Proceeding Series. Vol. 382 2009. 3.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Route kernels for trees
AU - Aiolli, Fabio
AU - Martino, Giovanni
AU - Sperduti, Alessandro
PY - 2009
Y1 - 2009
N2 - Almost all tree kernels proposed in the literature match substructures without taking into account their relative positioning with respect to one another. In this paper, we propose a novel family of kernels which explicitly focus on this type of information. Specifically, after defining a family of tree kernels based on routes between nodes, we present an efficient implementation for a member of this family. Experimental results on four different datasets show that our method is able to reach state of the art performances, obtaining in some cases performances better than computationally more demanding tree kernels.
AB - Almost all tree kernels proposed in the literature match substructures without taking into account their relative positioning with respect to one another. In this paper, we propose a novel family of kernels which explicitly focus on this type of information. Specifically, after defining a family of tree kernels based on routes between nodes, we present an efficient implementation for a member of this family. Experimental results on four different datasets show that our method is able to reach state of the art performances, obtaining in some cases performances better than computationally more demanding tree kernels.
UR - http://www.scopus.com/inward/record.url?scp=70049085909&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=70049085909&partnerID=8YFLogxK
U2 - 10.1145/1553374.1553377
DO - 10.1145/1553374.1553377
M3 - Conference contribution
AN - SCOPUS:70049085909
SN - 9781605585161
VL - 382
BT - ACM International Conference Proceeding Series
ER -