Abstract

We extend previous work on tree kernels to estimate the similarity between the dependency trees of sentences. Using this kernel within a Support Vector Machine, we detect and classify relations between entities in the Automatic Content Extraction (ACE) corpus of news articles. We examine the utility of different features such as Wordnet hypernyms, parts of speech, and entity types, and find that the dependency tree kernel achieves a 20 % F1 improvement over a “bag-of-words ” kernel. 1

Citations

6694	Statistical Learning Theory - Vapnik - 1998
2107	Introduction to statistical pattern recognition - Fukunaga - 1990
1491	Support-vector networks - Cortes, Vapnik - 1995
741	Shawe-Taylor J (2000) An Introduction to Support Vector Machines and other kernel-based learning methods - Cristianini
283	Extracting patterns and relations from the world wide web - Brin - 1998
282	Text classification using string kernels - Lodhi, Saunders, et al. - 2002
281	Convolutional kernels on discrete structures - Haussler - 2007
275	Snowball: Extracting relations from large plain-text collections - Agichtein, Gravano - 2000
204	Convolution kernels for natural languages - Collins, Duffy - 2001
130	Identity uncertainty and citation matching - Pasula, Marthi, et al. - 2002
78	A novel use of statistical parsing to extract information from text - Miller, Fox, et al. - 2000
73	Feature Engineering for Text Classification - Scott, Matwin - 1999
53	On kernel methods for relational learning - Cumby, Roth
42	Composite kernels for hypertext categorisation - Joachims, Cristianini, et al. - 2001