<?xml version="1.0" standalone="yes"?>
<Paper uid="W05-0633">
  <Title>Semantic Role Labeling Using Lexical Statistical Information</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> This paper presents a system for the CoNLL 2005 Semantic Role Labeling shared task (Carreras &amp; M`arquez, 2005), which is based on the current release of the English PropBank data (Palmer et al., 2005). For the 2005 edition of the shared task are available both syntactic and semantic information.</Paragraph>
    <Paragraph position="1"> Accordingly, we make use of both clausal, chunk and deep syntactic (tree structure) features, named entity information, as well as statistical representations for lexical item encoding.</Paragraph>
    <Paragraph position="2"> The set of features and their encoding reflect the necessity of limiting the complexity and dimensionality of the input space. They also provide the classifier with enough information. We explore here the use of a minimal set of compact features for semantic role prediction, and show that a feature-based statistical encoding of lexicalised features such as predicates, head words, local contexts and PoS by means of probability distributions provides an efficient way of representing the data, with the feature vectors having a small dimensionality and allowing to abstract from single words.</Paragraph>
  </Section>
class="xml-element"></Paper>