<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-2057">
  <Title>Sydney, July 2006. c(c)2006 Association for Computational Linguistics A FrameNet-based Semantic Role Labeler for Swedish</Title>
  <Section position="3" start_page="0" end_page="436" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Semantic role labeling (SRL), the process of automatically identifying arguments of a predicate in a sentence and assigning them semantic roles, has received much attention during the recent years.</Paragraph>
    <Paragraph position="1"> SRL systems have been used in a number of projects in Information Extraction and Question Answering, and are believed to be applicable in other domains as well.</Paragraph>
    <Paragraph position="2"> Building SRL systems for English has been studied widely (Gildea and Jurafsky, 2002; Litkowski, 2004), inter alia. However, all these works rely on corpora that have been produced at the cost of a large effort by human annotators. For instance, the current FrameNet corpus (Baker et al., 1998) consists of 130,000 manually annotated sentences. For smaller languages such as Swedish, such corpora are not available.</Paragraph>
    <Paragraph position="3"> In this work, we describe a FrameNet-based semantic role labeler for Swedish text. Since there was no existing training corpus available -- no FrameNet-annotated Swedish corpus of substantial size exists -- we used an English-Swedish parallel corpus whose English part was annotated with semantic roles using the FrameNet annotation scheme. We then applied a cross-language transfer to derive an annotated Swedish part. To evaluate the performance of the Swedish SRL system, we applied it to a small portion of the FrameNet example corpus that we translated manually. null</Paragraph>
    <Section position="1" start_page="0" end_page="436" type="sub_section">
      <SectionTitle>
1.1 FrameNet: an Introduction
</SectionTitle>
      <Paragraph position="0"> FrameNet (Baker et al., 1998) is a lexical database that describes English words using Frame Semantics (Fillmore, 1976). In this framework, predicates (or in FrameNet terminology, target words) and their arguments are linked by means of semantic frames. A frame can intuitively be thought of as a template that defines a set of slots, frame elements (FEs), that represent parts of the conceptual structure and typically correspond to prototypical participants or properties.</Paragraph>
      <Paragraph position="1"> Figure 1 shows an example sentence annotated with FrameNet information. In this example, the target word statements belongs to (&amp;quot;evokes&amp;quot;) the frame STATEMENT. Two constituents that fill slots of the frame (SPEAKER and TOPIC) are annotated as well.</Paragraph>
      <Paragraph position="2"> As usual in these cases, [both parties]SPEAKER agreed to make no further statements [on the  The initial versions of FrameNet were focused on describing situations and events, i.e. typically verbs and their nominalizations. Currently, however, FrameNet defines frames for a wider range of semantic relations that can be thought of as predicate/argument structures, including descriptions of events, states, properties, and objects.</Paragraph>
      <Paragraph position="3"> FrameNet consists of the following main parts: * An ontology consisting of a set of frames, frame elements for each frame, and relations (such as inheritance and causative-of) between frames.</Paragraph>
      <Paragraph position="4"> * A list of lexical units, that is word forms paired with their corresponding frames. The frame is used to distinguish between different senses of the word, although the treatment of polysemy in FrameNet is relatively coarsegrained. null * A collection of example sentences that provide lexical evidence for the frames and the corresponding lexical units. Although this corpus is not intended to be representative, it is typically used as a training corpus when contructing automatic FrameNet labelers.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>