<?xml version="1.0" standalone="yes"?>
<Paper uid="N06-1056">
  <Title>Learning for Semantic Parsing with Statistical Machine Translation</Title>
  <Section position="2" start_page="0" end_page="439" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Recent work on natural language understanding has mainly focused on shallow semantic analysis, such as semantic role labeling and word-sense disambiguation. This paper considers a more ambitious task of semantic parsing, which is the construction of a complete, formal, symbolic, meaning representation (MR) of a sentence. Semantic parsing has found its way in practical applications such as natural-language (NL) interfaces to databases (Androutsopoulos et al., 1995) and advice taking (Kuhlmann et al., 2004). Figure 1 shows a sample MR written in a meaning-representation language (MRL) called CLANG, which is used for ((bowner our {4}) (do our {6} (pos (left (half our))))) If our player 4 has the ball, then our player 6 should stay in the left side of our half.</Paragraph>
    <Paragraph position="1">  playing agents (Kuhlmann et al., 2004).</Paragraph>
    <Paragraph position="2"> Prior research in semantic parsing has mainly focused on relatively simple domains such as ATIS (AirTravelInformationService)(Milleretal., 1996; Papineni et al., 1997; Macherey et al., 2001), in which a typcial MR is only a single semantic frame. Learning methods have been devised that can generate MRs with a complex, nested structure (cf. Figure 1). However, these methods are mostly based on deterministic parsing (Zelle and Mooney, 1996; Kate et al., 2005), which lack the robustness that characterizes recent advances in statistical NLP. Other learning methods involve the use of fullyannotated augmented parse trees (Ge and Mooney, 2005) or prior knowledge of the NL syntax (Zettlemoyer and Collins, 2005) in training, and hence require extensive human efforts when porting to a new domain or language.</Paragraph>
    <Paragraph position="3"> In this paper, we present a novel statistical approach to semantic parsing which can handle MRs with a nested structure, based on previous work on semantic parsing using transformation rules (Kate et al., 2005). The algorithm learns a semantic parser given a set of NL sentences annotated with their correct MRs. It requires no prior knowledge of the NL syntax, although it assumes that an unambiguous, context-free grammar (CFG) of the target MRL is available. The main innovation of this al- null gorithm is its integration with state-of-the-art statistical machine translation techniques. More specifically, a statistical word alignment model (Brown et al., 1993) is used to acquire a bilingual lexicon consisting of NL substrings coupled with their translations in the target MRL. Complete MRs are then formed by combining these NL substrings and their translations under a parsing framework called the synchronous CFG (Aho and Ullman, 1972), which forms the basis of most existing statistical syntax-based translation models (Yamada and Knight, 2001; Chiang, 2005). Our algorithm is called WASP, short for Word Alignment-based Semantic Parsing. In initial evaluation on several real-world data sets, we show that WASP performs favorably in terms of both accuracy and coverage compared to existing learning methods requiring the same amount of supervision, and shows better robustness to variations in task complexity and word order.</Paragraph>
    <Paragraph position="4"> Section 2 provides a brief overview of the domains being considered. In Section 3, we present the semantic parsing model of WASP. Section 4 outlines the algorithm for acquiring a bilingual lexicon through the use of word alignments. Section 5 describes a probabilistic model for semantic parsing. Finally, we report on experiments that show the robustness of WASP in Section 6, followed by the conclusion in Section 7.</Paragraph>
  </Section>
class="xml-element"></Paper>