<?xml version="1.0" standalone="yes"?>
<Paper uid="C92-3129">
  <Title>THE IPS SYSTEM</Title>
  <Section position="4" start_page="0" end_page="0" type="metho">
    <SectionTitle>
3 The X module
</SectionTitle>
    <Paragraph position="0"> The central module of the IPS system is the module, which acts as the main generator of the system, and determines the syntactic structures of constituents. We assume the X schema in (1):</Paragraph>
    <Paragraph position="2"> where X is a lexical or a functional category, Spee and Compl are lists (possibly empty) of maximal projections (YP).</Paragraph>
    <Paragraph position="3"> As indicated in (1), maximal projections are of order 2 (XP) and lexicai categories of order 0 (I). For typographical reasons, categories of order 1 (~) are noted ' in the illustrations below. The set of lexical categories include N, V, * and p, the set of functional categories includes D(eterminer), T(ense) et C(omplementizer). We also assume the DP hypothesis (cf. Abney 1987, Clark 1990a), and, as a consequence, the strong parallelism between DP and TP structures, as illustrated in (2): (2)</Paragraph>
    <Paragraph position="5"> Lexical categories as well as functional cate+ gories can select other lexical or funrtional projectioos. Tiros, u determiner cyst select a projection of category Vp or NP, as in (3) and (4), respectively, corresponding to the structures (5)  and (6) (3) \[each, D, \[+definite\], \[__\[D,\[ .... rail\] ...... \] (4) \[each, D, \[+definite\], \[__\[N,\[singular\]\] ..... \]  (5)a, each five men.</Paragraph>
    <Paragraph position="6"> b. ivy \[D' each \[DP \[I)' five\[MP iN' menJJJl\]\] (6)a. each student.</Paragraph>
    <Paragraph position="7"> AcrEs DE COLING-92, NANTES, 23-28 AOlJq' 1992 8 7 1 PROC. OF COLING-92, NAbrrES. AUG. 23-28, 1992 b. \[ DP \[ D' each \[ NP I&amp;quot; N' student\]\]\]\] Similarly, auxiliaries can select projections of type UP, and most prepositions projections of type DP. Some examples of selection features associated with auxiliary verbs are given in (7), with the corresponding structures in (8) and (9):</Paragraph>
    <Paragraph position="9"> {have, V, \[+aux\], {__IV,{+ past par ticiplel\] .... \] \[be, V, \[+aux\], \[__\[V,\[past participle\]\] &amp;quot;a~\] \[be, V, \[+aux\], \[__IV,\[present participle\]\] .... \] the men have arrived.</Paragraph>
    <Paragraph position="11"> must \[Vp Iv' have \[Vp Iv' been \[Vp Iv' being \[ VP \[ V' cheated \[ DP e\]i\]\]\]\]\]\]\]\]\]\] The following is a summary of the fundamental properties of constituent structures :</Paragraph>
  </Section>
  <Section position="5" start_page="0" end_page="0" type="metho">
    <SectionTitle>
4 The IPS parsing strategy
</SectionTitle>
    <Paragraph position="0"> In our implementation of the X module, we distinguish three types of action: projection, attachment to the left (specifiers) and attachment to the riglit. The parsing strategy is left to right, parallel, combilting a bottom up approach with top down filtering, as we shall see below.</Paragraph>
    <Paragraph position="1"> The motivation for this particular strategy is to maximize at each step the interpretation of constituents in order to facilitate the selection mechanism as well as user interaction discussed in section 2. Interestingly enough, this requirement seems to coincide with psycholinguistic evidence suggesting that sentence parsing proceeds in an incremental fashion, trying to integrate incoming items into maximally interpreted structures ~.</Paragraph>
    <Paragraph position="2"> The basic idea is that the parser must be sensitive to incoming words. However, strictly bottom up strategies are known to have some undesirable features. In particular, they tend to generate numerous locally well-formed structures which turn out to be incompatible with the overall structure. Furthermore, they restrict attachment to complete constituents, which means that when applied to right branching languages such as French or English, assembling the final structure does not start much before the end of the sentence is reached.</Paragraph>
    <Paragraph position="3"> To illustrate these problems, consider the following examples : (10)a. Who could the children have invited ? b. John must have given the students several of his books.</Paragraph>
    <Paragraph position="4"> In sentence (10a), when the parser gets to the word have, it tries to combine it with the left context, say \[ DP the children\], leading to the new constituent \[ the children have\]. Al- TIP though this new constituent is perfectly well-formed locally, it is not compatible with the modal could.</Paragraph>
    <Paragraph position="5"> Sentence (10b) illustrates the second and more serious problem. If node attachment is limited to complete nodes, the combination of the subject John and the rest of the structure (the whole verb phrase, which is a * in our system) will not occur before the last word of the sentence is read.</Paragraph>
    <Paragraph position="6"> The use of a more sophisticated strategy, such as the left corner strategy, addresses the first problem quite successfully ~. However, it fails to solve the second problem, since attachments are limited to complete constituents in the standard left corner parser (~. In an attempt to overcome this problem, and taking advantage of the clear 4 For a detailed discussion attd review of the psycholinguistic evidence for incremental parsing see Tanenhaus et al. (1985), Frazier (1987) and Crocker (1992).</Paragraph>
    <Paragraph position="7"> SSee Aho and Unman (1972), Pereira and Shieber (1997) for a discussion of left corner parshig.</Paragraph>
    <Paragraph position="8"> 6Gibson (1991) proposes to relax this latter requirement~ and to allow attachments of incomplete constituents. However, the generality of his proposal still remains unclear.</Paragraph>
    <Paragraph position="9"> ACTES DE COLING-92, NANTES, 23-28 AoL~r 1992 8 7 2 PROC. OF COLING-92, NANTES, AUG. 23-28, 1992 bias for right branching structures in languages such as French or English (cf. structure (gb)), we developed a strategy dubbed &amp;quot;right corner&amp;quot;, which is based on the following principles: (11) &amp;quot;Right corner&amp;quot; strategy : constituents can be attached as soon as they have been hypothesized; attach inconfing items to maximally expanded constituents in the left context; constituents specify a list of active attachment sites on their right edge; all attachments are considered in parallel; Notice that tiffs strategy is clearly bottom-up (actions are triggered by incoming material). However, it differs from other bottom-up strate gies in some important ways. First of all, the combination of a new item with its left context is not done through a sequence of &amp;quot;reduce&amp;quot; operations (as in a shift-reduce parser). More generally, the attachment of (most) incoming items is made directly to some subconstituent of the top node, i.e. to some attachment site specified in the active node list of a constituent in the left context. This is in sharp contrast with standard bottom-up parsers (including the left corner parser), for which reduction to the start symbol cannot occur before the end of the sentence is reached.</Paragraph>
    <Paragraph position="10"> Although the right corner strategy requires significantly more complex data structures (constituents must specify all their potential attachment sitesT&amp;quot;), it has the advantage of being compntationally and psycholingulstically more adequate than other bottom up strategies. Regarding the latter point, the right corner strategy seems consistent with the (trivial) observations (i) that the analyzer is data driven but (it) although still incomplete, left context constituents are maximally specified, as they would in a top-down parser.</Paragraph>
    <Paragraph position="11"> A detailed example will illustrate tiffs strategy. Consider the following sentence : rAttachment sites for a given constituent correspond to the list of X nodea on its right edge. For efficiency reasons~ they are put together in a stack associated with the constituent.</Paragraph>
    <Paragraph position="12"> (12) Jotm has bought some flowers.</Paragraph>
    <Paragraph position="13"> When the parser reads tim word some, tile left context include, among many constituents, structure (13) : (13) \[ ~l' Johit has \[ Vl' boughtJJ Tile word some triggers a DP projection, with some as its head. Considering the left context, the parser raids structure (13), the stack of attachment sites of which contains the verb bought. The newly created DP projection combine with the TP structure as a complement of the verb bought. We now have an updated TP constituent, as in (14), with vm updated stack of active nodes, with at the top the DP constituent just attached.</Paragraph>
    <Paragraph position="14"> (14) I: &amp;quot;rP John has bought some\] The parser cent now read the following word flowers, which can attach to the left context structure (14), as a complement of the determiner some.</Paragraph>
    <Paragraph position="15"> The right corner strategy takes care of attachments to the right. In the case of projections or of attachments to the left (specifiers), the usual bottom up procedure applies. Typically projection is triggered by inherent features (\[+tense\] verbs will trigger a T(ense) projection, proper nouns a DP projection, etc.). As for left attachment, it occurs when the current constituent can find a left context constituent which can function as a possible specifier. The attachment of the specifier to the current constituent determines a new constituent which may in turn find a specifier in its left context (iterative attachmeat) as it happens in the possessive construction (e.g. John's little brother's cat).</Paragraph>
  </Section>
  <Section position="6" start_page="0" end_page="0" type="metho">
    <SectionTitle>
5 Concluding remarks
</SectionTitle>
    <Paragraph position="0"> Tile right coruer parsing strategy discussed in this paper has been developed to satisfy tile particular needs of our on-line interactive parsing model. By (i) pursuing concurrently all the possible analyses and (it) trying to integrate incoming items into fully developed constituents,</Paragraph>
  </Section>
class="xml-element"></Paper>