<?xml version="1.0" standalone="yes"?>
<Paper uid="P99-1053">
  <Title>Integrating multiple knowledge sources</Title>
  <Section position="3" start_page="413" end_page="415" type="intro">
    <SectionTitle>
3 Dialog Parsing
</SectionTitle>
    <Paragraph position="0"> The modifications required to a parser to implement this definition of dialog are relatively straightforward. At changes of  speaker, copies are made of all phrase hypotheses (arcs in a chart parser, for example) ending at the previous change of speaker. These copies are extended to the current change of speaker. We will use the term contribution (contr) here to refer to an uninterrupted sequence of words by one speaker (the words between speaker changes). In the example below, consider change of speaker (cos) 2. Copies of all phrase hypotheses ending at change of speaker 1 are extended to end at change of speaker 2. In this way, speaker A can form a phrase from contr-1 and contr-3 skipping speaker B's interruption, or contr-1, contr-2, and contr-3 can all form one constituent. At change of speaker 3, all phrase hypotheses ending at change of speaker 2 are extended to end at change of speaker 3 except those hypotheses that were extended from the previous change of speaker. Thus, an utterance cannot be formed from only contr-1 and contr-4. This mechanism implements the rules for speaker changes given in section 2: at each change of speaker, the new speaker can either build on the last contribution, build on their last contribution, or start a new utterance.</Paragraph>
    <Paragraph position="1"> A: contr-1 contr-3 B: contr-2 contr-4 cos 1 2 3 These rules assume that changes of speaker are well defined points of time, meaning that words of two speakers do not overlap. In the experiments of this paper, a corpus was used where word endings were time-stamped (word beginnings are unavailable). These times were used to impose an ordering; if one word ends before another it is counted as being before the other word. Clearly, this could be inaccurate given that words may overlap. Moreover, speakers may be slow to interrupt or may anticipate the first speaker and interrupt early. However, this approximation works fairly well as discussed in section 4.</Paragraph>
    <Paragraph position="2"> Other parts of the implementation are accomplished through metarules. The term metarule is used because these rules act not on words but grammar rules. Consider the editing term metarule. When an editing term is seen 2, the metarule extends copies of all phrase hypotheses ending at the editing term over that term to allow utterances to be formed around it. This metarule (and our other metarules) can be viewed declaratively as specifying allowable patterns of phrase breakage and interleaving (Core and Schubert, 1998). This notion is different from the traditional linguistic conception of metarules as rules for generating new PSRs from given PSRs. ~ Procedurally, we can think of metarules as creating new (discontinuous) pathways for the parser's traversal of the input, and this view is readily implementable. null The repair metarule, when given the hypothetical start and end of a reparandum (say from a language model such as (Heeman and Allen, 1997)), extends copies of phrase hypotheses over the reparandum allowing the corrected utterance to be formed. In case the source of the reparandum information gave a false alarm, the alternative of not skipping the reparandum is still available.</Paragraph>
    <Paragraph position="3"> For each utterance in the input, the parser needs to find an interpretation that starts at the first word of the input and ends at the last word. 4 This interpretation may have been produced by one or more applications of the repair metarule allowing the interpretation to exclude one or more reparanda. For each reparandum skipped, the parser needs to find an interpretation of what the user started to say. In some cases, what the user started to say is a complete constituent: take  editing terms might be via a metarule, X -&gt; Y Z ==&gt; X -&gt; Y editing-term Z, where X varies over categories and Y and Z vary over sequences of categories. However, this would produce phrases containing editing terms as constituents, whereas in our approach editing terms are separate utterances.</Paragraph>
    <Paragraph position="4"> 4In cases of overlapping utterances, it will take multiple interpretations (one for each utterance) to extend across the input.</Paragraph>
    <Paragraph position="5">  the oranges I mean take the bananas. Otherwise, the parser needs to look for an incomplete interpretation ending at the reparandum end. Typically, there will be many such interpretations; the parser searches for the longest interpretations and then ranks them based on their category: UTT &gt; S &gt; VP &gt; PP, and so on. The incomplete interpretation may not extend all the way to the start of the utterance in which case the process of searching for incomplete interpretations is repeated. Of course the search process is restricted by the first incomplete constituent. If, for example, an incomplete PP is found then any additional incomplete constituent would have to expect a PP.</Paragraph>
    <Paragraph position="6"> Figure 1 shows an example of this process on utterance 62 from TRAINS dialog d92a-</Paragraph>
    <Section position="1" start_page="415" end_page="415" type="sub_section">
      <SectionTitle>
1.2 (Heeman and Allen, 1995). Assuming
</SectionTitle>
      <Paragraph position="0"> perfect speech repair identification, the repair metarule will be fired from position 0 to position 5 meaning the parser needs to find an interpretation starting at position 5 and ending at the last position in the input.</Paragraph>
      <Paragraph position="1"> This interpretation (the corrected utterance) is shown under the words in figure 1. The parser then needs to find an interpretation of what the speaker started to say. There are no complete constituents ending at position 5. The parser instead finds the incomplete constituent ADVBL -&gt; adv * ADVBL.</Paragraph>
      <Paragraph position="2"> Our implementation is a chart parser and accordingly incomplete constituents are represented as arcs. This arc only covers the word through so another arc needs to be found.</Paragraph>
      <Paragraph position="3"> The arc S -&gt; S * ADVBL expects an ADVBL and covers the rest of the input, completing the interpretation of what the user started to say (as shown on the top of figure 1). The editing terms are treated as separate utterances via the editing term metarule.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>