<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-3014">
  <Title>Parsing and Subcategorization Data</Title>
  <Section position="3" start_page="0" end_page="79" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Robust statistical syntactic parsers, made possible by new statistical techniques (Collins, 1999; Charniak, 2000; Bikel, 2004) and by the availability of large, hand-annotated training corpora such as WSJ (Marcus et al., 1993) and Switchboard (Godefrey et al., 1992), have had a major impact on the field of natural language processing. There are many ways to make use of parsers' output. One particular form of data that can be extracted from parses is information about subcategorization. Subcategorization data comes in two forms: subcategorization frame (SCF) and sub-categorization cue (SCC). SCFs differ from SCCs in that SCFs contain only arguments while SCCs contain both arguments and adjuncts. Both SCFs and SCCs have been crucial to NLP tasks. For example, SCFs have been used for verb disambiguation and classification (Schulte im Walde, 2000; Merlo and Stevenson, 2001; Lapata and Brew, 2004; Merlo et al., 2005) and SCCs for semantic role labeling (Xue and Palmer, 2004; Punyakanok et al., 2005).</Paragraph>
    <Paragraph position="1"> Current technology for automatically acquiring subcategorization data from corpora usually relies on statistical parsers to generate SCCs. While great efforts have been made in parsing written texts and extracting subcategorization data from written texts, spoken corpora have received little attention. This is understandable given that spoken language poses several challenges that are absent in written texts, including disfluency, uncertainty about utterance segmentation and lack of punctuation. Roland and Jurafsky (1998) have suggested that there are substantial subcategorization differences between written corpora and spoken corpora. For example, while written corpora show a much higher percentage of passive structures, spoken corpora usually have a higher percentage of zero-anaphora constructions. We believe that sub-categorization data derived from spoken language, if of acceptable quality, would be of more value to NLP tasks involving a syntactic analysis of spoken language, but we do not pursue it here.</Paragraph>
    <Paragraph position="2"> The goals of this study are as follows:  1. Test the performance of Bikel's parser in parsing written and spoken language.</Paragraph>
    <Paragraph position="3"> 2. Compare the accuracy level of SCCs gen- null erated from parsed written and spoken language. We hope that such a comparison will shed some light on the feasibility of acquiring SCFs from spoken language using the cur- null rent SCF acquisition technology initially designed for written language.</Paragraph>
    <Paragraph position="4"> 3. Explore the utility of punctuation1 in parsing and extraction of SCCs. It is generally recognized that punctuation helps in parsing written texts. For example, Roark (2001) finds that removing punctuation from both training and test data (WSJ) decreases his parser's accuracy from 86.4%/86.8% (LR/LP) to 83.4%/84.1%. However, spoken language does not come with punctuation. Even when punctuation is added in the process of transcription, its utility in helping parsing is slight. Both Roark (2001) and Engel et al. (2002) report that removing punctuation from both training and test data (Switchboard) results in only 1% decrease in their parser's accuracy.</Paragraph>
  </Section>
class="xml-element"></Paper>