Semantic Role Labeling Using a Grammar-Driven Convolution Tree Kernel Export

@article{ZhangEtAl:08, abstract = {Convolution tree kernel has shown promising results in semantic role labeling (SRL). However, this kernel does not consider much linguistic knowledge in kernel design and only performs hard matching between subtrees. To overcome these constraints, this paper proposes a grammar-driven convolution tree kernel for SRL by introducing more linguistic knowledge. Compared with the standard convolution tree kernel, the proposed grammar-driven kernel has two advantages: 1) grammar-driven approximate substructure matching, and 2) grammar-driven approximate tree node matching. The two approximate matching mechanisms enable the proposed kernel to better explore linguistically motivated structured knowledge. Experiments on the CoNLL-2005 SRL shared task and the PropBank I corpus show that the proposed kernel outperforms the standard convolution tree kernel significantly. Moreover, we present a composite kernel to integrate a feature-based polynomial kernel and the proposed grammar-driven convolution tree kernel for SRL. Experimental results show that our composite kernel-based method significantly outperforms the previously best-reported ones.}, author = {Zhang, Min and Che, Wanxiang and Zhou, Guo D. and Aw, Aiti and Tan, Chew L. and Liu, Ting and Li, Sheng}, booktitle = {Audio, Speech, and Language Processing, IEEE Transactions on}, citeulike-article-id = {4669424}, citeulike-linkout-0 = {http://dx.doi.org/10.1109/TASL.2008.2001104}, citeulike-linkout-1 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4595687}, doi = {10.1109/TASL.2008.2001104}, journal = {Audio, Speech, and Language Processing, IEEE Transactions on}, number = {7}, pages = {1315--1329}, posted-at = {2009-08-17 13:15:45}, priority = {2}, title = {Semantic Role Labeling Using a Grammar-Driven Convolution Tree Kernel}, url = {http://dx.doi.org/10.1109/TASL.2008.2001104}, volume = {16}, year = {2008} }

TY - JOUR ID - ZhangEtAl:08 L3 - citeulike-article-id:4669424 N2 - Convolution tree kernel has shown promising results in semantic role labeling (SRL). However, this kernel does not consider much linguistic knowledge in kernel design and only performs hard matching between subtrees. To overcome these constraints, this paper proposes a grammar-driven convolution tree kernel for SRL by introducing more linguistic knowledge. Compared with the standard convolution tree kernel, the proposed grammar-driven kernel has two advantages: 1) grammar-driven approximate substructure matching, and 2) grammar-driven approximate tree node matching. The two approximate matching mechanisms enable the proposed kernel to better explore linguistically motivated structured knowledge. Experiments on the CoNLL-2005 SRL shared task and the PropBank I corpus show that the proposed kernel outperforms the standard convolution tree kernel significantly. Moreover, we present a composite kernel to integrate a feature-based polynomial kernel and the proposed grammar-driven convolution tree kernel for SRL. Experimental results show that our composite kernel-based method significantly outperforms the previously best-reported ones. IS - 7 JF - Audio, Speech, and Language Processing, IEEE Transactions on EP - 1329 TI - Semantic Role Labeling Using a Grammar-Driven Convolution Tree Kernel VL - 16 T2 - Audio, Speech, and Language Processing, IEEE Transactions on SP - 1315 AU - Zhang, Min AU - Che, Wanxiang AU - Zhou, Guo AU - Aw, Aiti AU - Tan, Chew AU - Liu, Ting AU - Li, Sheng PY - 2008/// UR - http://dx.doi.org/10.1109/TASL.2008.2001104 ER -