<?xml version="1.0" standalone="yes"?>
<Paper uid="W99-0104">
  <Title>O @ O O O O O O O O O O O @ O O O 0 O O O O O O O @ O 0 O @ O O O O @ O O O O O @ O @ O</Title>
  <Section position="3" start_page="29" end_page="33" type="intro">
    <SectionTitle>
2 Coreference Resolution
</SectionTitle>
    <Paragraph position="0"> Coreference resolution relies on a combination of linguistic and cognitive aspects of language. Linguistic constraints are provided mostly by the syntactic modeling of language, whereas computational models of discourse bring forward the cognitive aesumplions of anaphora resolution. Three different methods of combining anaphoric constraints am known to date. The Rrst one integrates anaphora resolution in computational models of discourse interpretation.</Paragraph>
    <Paragraph position="1"> Dynamic properties of discourse, especially focusing and centering are invoked as the primary b~-~|~ for identifying antecedents. Such computational methods were presented in (Grosz et al., 1995) and (Webber, 1988).</Paragraph>
    <Paragraph position="2"> A second category of approaches combines a v~ riety of syntactic, semantic and discourse factors as a multi-dimensional metric for ranking antecedent candidates. Anaphora resolution is determined by a composite of several distinct scoring procedures, each of which scores the prominence of the candidate with respect to a specific.type of information. The systems described in (Asher and Wada, 1988) (Carbonell and Brown, 1988) and (Rich and Luperfoy, 1988) are examples of the mixed evaluation strategy. null Alternatively, other discourse-based methods consider co~eference resolution a by-product of the recognition of coher~ce relations between sentences.</Paragraph>
    <Paragraph position="3"> Such methods were presented in (Hoblm et al., 1993) and ~flensky, 1978). Although M-complete, this approach has the appeal that it resolves the most complicated cases of coreference, uncovered by syntactic or semantic cues. We have revisited these methods by setting the relation between coreference and coherence on empirical grounds.</Paragraph>
    <Section position="1" start_page="29" end_page="31" type="sub_section">
      <SectionTitle>
2.1 Pronominal Coreference
</SectionTitle>
      <Paragraph position="0"> Two tendencies characterize current pronominal coreference algorithms. The first one makes use of the advances in the parsing technology or on the availability of large parsed corpora (e.g. Trcebank (Marcus et al.1993)) to produce algorithms inspired by Hobbs' baseline method (Hobbs, 1978). For example, the Resolution of Anaphor~ Procedure (RAP) i~itroduced in (Lappin and Leass, 1994) combines syntactic information with agreement and salience constraints. Recently, a probabilistic approach to pronominal coreference resolution was also devised (Ge et al., 1998), using the parsed data available from Treebank. The knowledge-based method of Lappin and Leass produces better results. Nevertheless, RkPSTAT, a version of RAP obtained by using statistically measured preference patterns for the antecedents, prodticed a slight enhancement of performance over RAP.</Paragraph>
      <Paragraph position="1"> Other pronominal resolution approaches promote knowledge-poor methods (Mitkov , 1998), either by using an ordered set of general heuristics or by combining scores assigned to candidate antecedents.</Paragraph>
      <Paragraph position="2"> The CogNIAC algorithm (Baldwin, 1997) uses six heuristic rules to resolve coreference, whereas the algorithm presented in (Mitkov, 1998) is based on a limited set of preferences (e.g. definitiveness, lexical reiteration or immediate reference). Both these algorithm rely only on part-of-speech tagging of texts and on patterns for NP identification. Their performance (dose to 90% for certain types of pronouns) indicates that full syntactic knowledge is not required by certain forms of pronominal coreference.</Paragraph>
      <Paragraph position="3"> The same claim is made in (Kennedy and Boguraev, 1996) and (Kameyama, 1997), where algorithm~ approximating RAP for poorer syntactic input obtain precision of 75% and 71%, respectively, a surprising small precision decay from RAP's 86%. These results prompted us to devise COCKTAIL, a corderence resolution system, as a mixture of heuristics performing on the various syntactic, semantic and dL~ourse cues. COCKTAIL is a composite of heuristics learned from the tagged corpora, which has the following novel characteristics: 1. C0cErIIL covers both nominal and pronoun corder~ce, but distinct sets of heuristics operate for different forms of anaphors. We have devised separate heuristics for reflexive, possessive, relative, 3rd person and 1st person pronouns. Similarly, de/inite nomlo-t~ are treated differently than bare or inddinite nominals. 2. c0crr/IL performs semantic checks between antecedents and ~phorL These chedm combine sottal co~aints from WordNet with co-occurance information from (a) Treebank and (b) conceptual  3. In COCET~L antecedents are sought not only in the ac~e~ble text region, but we also throughout the current co~efe~nce chains. In this way cohesive in- null formation, represented in corderence chains, is employed i~ the resolution pr _~___. 4. The heuristics d ~cErAIL allow for lexi~dizations (e.g. when the anaphor is an adjunct ofa commmdcation verbs) and of simplified coherence cues (e.g.  V when the anaphor is the subject of verb add, the antecedent may be a preceding subject of a communication vehb).</Paragraph>
      <Paragraph position="4"> To exemplify some COCKTAIL heuristics that resolve pronominal coreference, we first present heuristics applicable for reflexive pronoun and then we list heuristics for possessive pronouns and 3rd person pronoun resolution. Brevity imposes the omission of heuristics for other forms of pronoun resolution.</Paragraph>
      <Paragraph position="5"> COCKTAIL operates by successively applying the following heuristics to the pronoun Pro~ Oif (Pron is reflezive) then apply successively: oHenristic 1-Reflexive(H1R) Search for PN, the closest proper name from Prgn in the same sentence, in right to left order.</Paragraph>
      <Paragraph position="6"> if (PN agrees in number and gender with Pron) if (PN belongs to core/erence chain CC) then Pick the element from CC which is closest to Pron in Text.</Paragraph>
      <Paragraph position="7"> else Pick PN.</Paragraph>
      <Paragraph position="8"> o Henr/stic 2-Refle~'ve(H2R) Search for a sequence Noun.Relative.Pronoun, in the same sentence, in rigld to left order.</Paragraph>
      <Paragraph position="9"> if (Noun agrees in number and gender with Pron) if (Noun belongs to earefe~n~ chain CC) then Pick the dement from CC which is closest to Iron in Tezt.</Paragraph>
      <Paragraph position="10"> else Pick Noun.</Paragraph>
      <Paragraph position="11"> oHeur/sHc $-Refle~'/~e(H3R) Search for Pron&amp;quot; the closest prenoun from Pron in the same sentence, in right to left order.</Paragraph>
      <Paragraph position="12"> if (Pron&amp;quot; agrees in number and gender with Pron) if (Pron' ~gs to wreferen~ chain CC) then Pick the dement from CC which is closest to Pron in Tcet.</Paragraph>
      <Paragraph position="13"> eLse Pick Pron: o Heuristie 4-Reflezive(H4R) Search/or Noun.e, the dosest noun .from Pron in the same sentence, in right to left order.</Paragraph>
      <Paragraph position="14"> if (Noun.c a#n~a in number and gender with Pron) then Pick Noun.~ Resolution examples for reflexive pronouns are illustrated in Table L The antecedents produced by COCKTAIL are boldfaced, whereas the referring expressions are emphasized. Both referring expressions and resolved antecedents and underlined. Precision results are listed in Table 2.</Paragraph>
      <Paragraph position="15"> Antecedents of reflexive pronouns are always sought in the same sentence. Antecedents of other types of pronouns are sought in preceding sentences too, starting from the immediately preceding sentence. Inside the sentence, the search for a specific word is performed from the current position towards the beginning of the sentence, whereas in the pre-Before Pennzoii's court fight with Texaco over the Getty purchase, Mr. Liedtke - one of the ploy's foremost practitioners - portrayed him.~elfas something of an oil-patch tube, a notable f~---'~&amp;quot;~-~nsidering his diplomas from Amherst College and Harvard Business School.</Paragraph>
      <Paragraph position="16"> The' woman who is kuown to me as hard-working and. responsible, clearly isn't hersel/.</Paragraph>
      <Paragraph position="17"> Unlike many of her peers, m~t of whom are males in their 30s, s.he never takes herself too seriously.  ceding sentences, the search starts at the beginning of the sentence and proceeds in a left to right fashion. The same search order was used in (Kameyama, 1997). From now on, we indicate this search by Searchl. This search is employed by heuristics for possessive pronoun resolution: Oif (Pron is possessive) (i.e. we have a sequence \[Pron nouno\], where nouno is the head of the NP containing Pron) then apply suco,~_~sieely: o Henris6C/. l-Pouessive(H IPos ) Searchl /or a posses~ve comb'uct of the form \[.ounl's ,~n2\], if (\[Pron nouno\] and \[nounl's noun2\] agree in gender, n-tuber and are semantically consistent) then if (noun2 belonga to coreyerence chain CC) and there is andement from CC which is closest to Pron in Tezt, Pick that dement.</Paragraph>
      <Paragraph position="18"> Pick noun,.</Paragraph>
      <Paragraph position="19"> oHcur/sfc 2-pouess/ve(H2Pos) Senrchl for PN, the closest proper name from Pron if (PN agrees in number and gender with Pron) if (PN belongs to corefe~n~ chain CC) then ~ the dement from CC ~hich is closest to Pron in Tezt else Pick PN.</Paragraph>
      <Paragraph position="20"> oHeuris6c 3-Possessive(H3Pce).</Paragraph>
      <Paragraph position="21"> Search for Pron&amp;quot; the closest pronoun .from Pron . if (Pron&amp;quot; egre~ in number and fender e~h Pron). if (Pron' belongs to coreferen~ chain CC) and there is an dement from CC which is closest to Pron in Text, Pick that element.</Paragraph>
      <Paragraph position="22"> else Pick Pron' oHenrist~ ~.Possessiee(H4Pos) Search for Noun, the closest eammon noun from Iron if (Noun agrees in number and gender with Pron)  if (Noun belongs to coref~ chain CC) and there is an element from CC which is closes~ to Pron in Tezt, Pick that element.</Paragraph>
      <Paragraph position="23"> else Pick Noun Examples and precision results are listed in Table 3 and Table 4, respectively.</Paragraph>
      <Paragraph position="24"> The timing of Mr. Shad's departure is likely to depend on how rapidly the Senate Banking Committee moves to confirm his successor.</Paragraph>
      <Paragraph position="25"> Ronald Reagan sends him-a list of h/s film roles. The 20-minute tiigfit )~elps him forget h/s troubles. The president renewed h/s promise to veto  Given a possessive pronotm in a sequence \[Pron Noon0\], the antecedent Ante of Pron is semanti. cal\]y consistent if the same possessive relationship can be established between Ante and Noono. the problem is that the possessive relation semantically corresponds to an open list of relations. For example, Nouno may be a feature of Ante. Ante may own Noono or Ante may have pe, formed the action lexical/zed by the nominali~-~on Nouno.</Paragraph>
      <Paragraph position="26"> COCKTAIL's test of semantic consistency blends togerber information available from WordNet and on statistics gathered from ~ebank. Different consistency checks are modeled for each of the heuristics. We detail here the check that applies to heuristic HIPos, that resolves the possessive from the first example listed in Table 3. For this heuristic, we have to test whether from the possessive \[Ante Nount\] we can grant the pos~_~ve \[Ante Noone\] as well. There axe three cases that allow us to do so: * ~ase 1 Nount and Nouno corder.</Paragraph>
      <Paragraph position="27"> * Case ~Theceis ase~se ss of Nounx and asense so of Nouno such that a synonym of Noun~ i or of its immediate hypernym is found in the gloss of Noon~ or vicevers&amp; * ~ There is a sense st of Nounx and a sense So of Nouno such that a common concept is found in their glosses.</Paragraph>
      <Paragraph position="28"> Cases 2 and 3 extend to synsets obtained through derivational morphology as well (e.g. nominalizations). For cases 2 and 3 COCKTAIL reinforces the coreference hypothesis by using a possessive. similarity metric based on Resuik's similarity measures for noun groups (B___,~m_ i_k, 1995). From a subset of Treebank, we collect all possessives, and measure whether the similarity~clam of Nouno, Noun1 and their eventual common concept is above a threshold produced off-line.</Paragraph>
      <Paragraph position="29"> Other pronominal coreference heuristics employ Search2, a search procedure that enhances Searchx, since it prefers antecedents that are immediately succeeded by relative pronouns. This search is in. corporated in COCKTAIL's heuristics that resolve 3rd person pronominal coreference: o Heuristic 1-Prono.un_(HIPron) Search2 in the same sentence for the same 5rd person pronoun Pron' if (Pron' belongs to coreference chain CC) and there is an element from CC which is closest to Pron in Text, Pick that dement.</Paragraph>
      <Paragraph position="30"> else Pick Pron&amp;quot; oHeuristic ~-Prenoon(H2Pron) Search2 for PN, the closest proper name from Pron if (PN agrees in number and gender with'Pron) if (PN belongs&amp;quot; to coreference chain CC)  then Pick the element from CC which is closest to Pron in. Text.</Paragraph>
      <Paragraph position="31"> else Pick PN.</Paragraph>
      <Paragraph position="32"> oHeuristic 3-Prenoon(H3Pron) if Pron collocates with a communication verb then Searcht for pronoon Pron'--I if (Pron&amp;quot; belongs to C/oreference chain CC) and there is an clement from CC e~hich is closest to Iron in Tezt, Pick that dement else Pick Pron&amp;quot; oHeuristic ~-Pronoun(H4Pron) if Pron collocates with a communication verb thell Search\] communicator Noun if (#oun belongs to coreyeren~ chain CC) and there im an clement from CC u#sich is clos/Jt to Pmn in Te.zt, Pick that dement.</Paragraph>
      <Paragraph position="33"> else Pick Noo~ o Heuristic 5-Pmnoon(HSPron) . Searcha for Pron', the closest pronoun from Pron  if (Pron' agrees in number and gender with Pron) if (Pron' belonga to C/oneference chain CC) and there is an dement from CC tnhich is do, eat to Pron in Teffit, Pick that dement else Pick Pren&amp;quot; oHfu~ 6-Proooen(H6Pron) Search2 for Noun, the closest noun from.Pron if (Noun agrees in number and gender with Pron) if (Noon belongs to coreferen~ chain CC) and there is an element from CC which is dosest to Iron in Tezt, Pick that dement.</Paragraph>
      <Paragraph position="34">  olution. FYom our initial experiments, we do not see the need for special semantic consistency checks, since all heuristics performed with precision in excess of 90% Part of this is explained by our usage of pleonastic filters and of recognizers of idiomatic usage. Table 5 illustrates some of the successful coreference resolutions.</Paragraph>
      <Paragraph position="35"> H_qe says that in many years as a banker he has grown accustomed to &amp;quot;dealing with honest people 99% of the time.</Paragraph>
      <Paragraph position="36"> sen. Byrd takes pains to reassure the voter that he will see to it that the trade picture improves. A..nurse who deals with the new patient ~Jmits sh.._~e isn't afraid of her temper.</Paragraph>
    </Section>
    <Section position="2" start_page="31" end_page="33" type="sub_section">
      <SectionTitle>
2.2 Nominal Coreference
</SectionTitle>
      <Paragraph position="0"> Noun phrases can represent referring expressions in a variety of cases. For example, it is known that not all definite NPs are anaphoric. Conditions that define anaphoric NPs are still under research (cf.</Paragraph>
      <Paragraph position="1"> (Poesio and Vieira, 1998)). In the tagged corpora, we have found only 20.93% of the nominal coreference cases to be definites, the majority (78.85%) being bare nominals 2, and only 1.32% were inclefiuites.</Paragraph>
      <Paragraph position="2"> However, more than 50% of the nominal referring expressions were names of people, org~n!-~tions or locations. Adding to this, 15.22% of nominal coreference links are accounted by appositives. Based on this evidence, COCKTtIL implements special rules for name alias identification and for robust recognition of appositions. Moreover, the heuristics for nominal coreference resolution apply Senrchs, and enhancement of Search~ that searches starting with the coreference chains, and then with the accessible text. To resolve nominal coref~eace, COCKTAIL successively applies the following heuristics: oHeuristic l.Nominal(H1Nom) if (Noun is the head of an appositive) then Pick the preceding NP.</Paragraph>
      <Paragraph position="3"> o Heuristic P..Nor~inal(H2Nom) if (Noun belongs to an NP, Searchs /or NP' such that Noun'ffiaame_name(head(NP),head(NP')) or Noun'--same.name(adj(NP),adj(Ne'))) then if (Noun' belongs to core/erence chain CO) then Pick the element ~vm CC which is closest to Noun in Text.</Paragraph>
      <Paragraph position="4"> else Pick Noun: oHeuristif. 3-Nominal(H3Nom) if Noun is the head of an NP then Searchs for proper name PN 2We count as bare nominals coreferring adjuncts as well. such that head(PN)-Noun if (PN belongs to coreference chain CG) and there is an element from CC which is closest to Noun in Text, Pick that element.</Paragraph>
      <Paragraph position="5"> else Pick PN.</Paragraph>
      <Paragraph position="6"> o Houristie 4-Nominal( H 4N om) Searchs \]or a proper name PN with the same category as Noun if (PN belongs to core-ference chain CC) and there is an element from CO which is closest to Noun in Tezt, Pick that element.</Paragraph>
      <Paragraph position="7"> else Pick PN.</Paragraph>
      <Paragraph position="8"> oHeuristic 5-Nominai(H5Nom) Searchs Noun&amp;quot; a spnenym or hyponyrn of Noun if (Noun' belongs to core/erence chain CC) and there is an element fl'om CO which is closest to Noun in Text, Pick that dement.</Paragraph>
      <Paragraph position="9"> else Pick Noun'.</Paragraph>
      <Paragraph position="10"> oH. euristic 6-Nominal(H6Nom ) Searchs for Noun either in definites or in NPs having adjuncts in coreyerence chain CU) if Ante 8emantieally consistent with Noun if (Ante belongs to core/erenee chain UC) and there is an dement from UU which is closest to Noun in Text, Pick that element:  else Pick Ante.</Paragraph>
      <Paragraph position="11"> oHeuristic 7-Nomine/(H7Nom) if (Noun or one ol his hz~n~nrts or holonyms is a nominalization N) then Search/or the verb V deriving N or one o/ its synen~ns) then P/ok NP, the closest adjunct o/V if (NP belongs to C/ore!erence chain 00) az~d there is an dement from CO which is closest to Noun in Te~, Pick that element.</Paragraph>
      <Paragraph position="12"> else Pick NP</Paragraph>
      <Paragraph position="14"> if (Noun is the head o/a prepositional phrase preceded by a nominalization N) then Search/or the verb V deriving N or one oI its s~um~ns) if (Noun&amp;quot; is on adjunct o/ V) and (Noun&amp;quot; and Noun have the same category * if (Noun' belongs to C/ore/erenea chain CC) and there is an dement from CC which is closest to ~Voen in Text, Pick that dement~  me o p. es l by appositions, whereas heuristic H2Nom promotes  IMB and Mr. York would;t discuss his compensation package which could easily reach into seven figures. ~ect is sensitive at a time when IMB is !aying off thousands of employees Mr Iacocca led Chrysler through one of the 'largest stock sales ever for a U.S. industrial company, raising .$1.78 billion. Chrysler is using most of the proceeds to reduce its $4.4. billion unfunded pension liability. We read where the Clinton White House is seeking a deputy to chief of staff Mack McLarty to impose some disciplined coherence on the p/ace's * ambunctious young staff.</Paragraph>
      <Paragraph position="15">  the term repetition indicator, when consistency checks apply. For this heuristic, consistency checks are conservative, imposing that either the adjuncts be identical, coreferring or the adjunct of the referent be less specific than the antecedent. Specificity principles apply also to HSNom, where hyponymy is promoted, similarly to (Poesio and Vieirs, 1998). Heuristic H3Nom allows coreference between &amp;quot;the Securities and F_,z~ange Commission n and .~he commission ~ but it bans links between ~Reardon Steel Co.&amp;quot; and &amp;quot;tons of steal&amp;quot;. Many times coreferring nomln~l~ share a~o semantic relations (e.g. synonym#). Heuristic HSNom identifies such cases, by applying consistency checks. Based on experiments with the coreference module of FASTUS, where this heuristic was initially implemented, we require that most frequent senses of nouns be promoted. The same precedence of f~quent senses is implemented in the assi~ment of categories, defined as the immediate WordN~ h~ pernTpn. The category of proper names is dictated by the proper name recognizer, ~qlo~ing such categories m Person, Organization or In this way, coreference between &amp;quot;IBM ~ and ~he wo,mded computer 9lent ~ can be estab!|~bed, since sense 3 of noun #/ant is Organim6on, the category of ~IBM~. Simi!m- ~tegory-based semaatic cheCkS allow the recognition of the antecedent of proceeds from the second example listed in Table 6. The h~l~ern~ of ~eceezk is ga/n, whose glou genus is amount, the category of $1.78 biUio~ Semantic checks are also required in H?Nom and HSNom, heuristic that rely on derivational morphology. The first example from Table 6 is resolved by HTNom, since d/scass/on the nominalization of d/scuss b~_q the category communication, a hypernym of subject, The antecedent is the object of the verb d/scuss. The last heuristic, H9Nom identifies coreferring links with coerced entities of nominals. Coercions are obtained as paths of meronyms or hypernyms.</Paragraph>
      <Paragraph position="16"> (Harabagiu, 1998) discusses a coercion methodology based on WordNet and Treebank. Since in our test corpus there we very few cases of metonymic anaphors, Table 7 lists the precision of the other heuristics only.</Paragraph>
      <Paragraph position="17">  The empirical * methods employed in COCKTAIL are an alternative to the inductive approaches described in (Cardie and Wagstatf, 1999) and (McCarthy and Lehnert, 1995). Our results show that high-precision empirical techniques can be ported from pronominal coreference resolution to the more difficult problem of nominal coreference.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>