<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-2010">
  <Title>Sydney, July 2006. c(c)2006 Association for Computational Linguistics A Hybrid Convolution Tree Kernel for Semantic Role Labeling</Title>
  <Section position="4" start_page="73" end_page="73" type="relat">
    <SectionTitle>
2 Related Work
</SectionTitle>
    <Paragraph position="0"> Automatic semantic role labeling was first introduced by Gildea and Jurafsky (2002). They used a linear interpolation method and extract features from a parse tree to identify and classify the constituents in the FrameNet (Baker et al., 1998) with syntactic parsing results. Here, the basic features include Phrase Type, Parse Tree Path, Position.</Paragraph>
    <Paragraph position="1"> Most of the following works focused on feature engineering (Xue and Palmer, 2004; Jiang et al., 2005) and machine learning models (Nielsen and Pradhan, 2004; Pradhan et al., 2005a). Some other works paid much attention to the robust SRL (Pradhan et al., 2005b) and post inference (Punyakanok et al., 2004).</Paragraph>
    <Paragraph position="2"> These feature-based methods are considered as the state of the art method for SRL and achieved much success. However, as we know, the standard flat features are less effective to model the syntactic structured information. It is sensitive to small changes of the syntactic structure features. This can give rise to a data sparseness problem and prevent the learning algorithms from generalizing unseen data well.</Paragraph>
    <Paragraph position="3"> As an alternative to the standard feature-based methods, kernel-based methods have been proposed to implicitly explore features in a high-dimension space by directly calculating the similarity between two objects using kernel function. In particular, the kernel methods could be effective in reducing the burden of feature engineering for structured objects in NLP problems. This is because a kernel can measure the similarity between two discrete structured objects directly using the original representation of the objects instead of explicitly enumerating their features.</Paragraph>
    <Paragraph position="4"> Many kernel functions have been proposed in machine learning community and have been applied to NLP study. In particular, Haussler (1999) and Watkins (1999) proposed the best-known convolution kernels for a discrete structure. In the context of convolution kernels, more and more kernels for restricted syntaxes or specific domains, such as string kernel for text categorization (Lodhi et al., 2002), tree kernel for syntactic parsing (Collins and Duffy, 2001), kernel for relation extraction (Zelenko et al., 2003; Culotta and Sorensen, 2004) are proposed and explored in NLP domain. Of special interest here, Moschitti (2004) proposed Predicate Argument Feature (PAF) kernel under the framework of convolution tree kernel for SRL. In this paper, we follow the same framework and design a novel hybrid convolution kernel for SRL.</Paragraph>
  </Section>
class="xml-element"></Paper>