Resolving Individual and Abstract Anaphora in Texts and
Dialogues
Costanza Navarretta
Center for Sprogteknologi, University of Copenhagen
Njalsgade 80
2300 Copenhagen S
costanza@cst.dk
Abstract
This paper describes an extension of the
dar-algorithm (Navarretta, 2004) for re-
solving intersentential pronominal anaphors
referring to individual and abstract enti-
ties in texts and dialogues. In dar indi-
vidual entities are resolved combining mod-
els which identify high degree of salience
with high degree of givenness (topicality)
of entities in the hearer-s cognitive model,
e.g. (Grosz et al., 1995), with Hajiˇcov´aet
al.-s (1990) salience account which assigns
the highest degree of salience to entities
in the focal part of an utterance in Infor-
mation Structure terms, which often intro-
duce new information in discourse. Anaph-
ors referring to abstract entities are resolved
with an extension of the algorithm pre-
sented by Eckert and Strube (2000). The
extended dar-algorithm accounts for diﬀer-
ences in the resolution mechanisms of diﬀer-
ent types of Danish pronouns. Manual tests
of the algorithm show that dar performs
better than other resolution algorithms on
the same data.
1 Introduction
Although most pronominal anaphora resolu-
tion algorithms only account for anaphors re-
ferring to individual entities, anaphors re-
ferring to abstract entities evoked by verbal
phrases, clauses or discourse segments (hence-
forth APAs) are quite common in English see
i.a. (Byron and Allen, 1998) and even more
in Danish (Navarretta, 2004). Recently, two
algorithms for resolving APAs and individ-
ual pronominal anaphors (henceforth IPAs) in
specific English dialogues have been proposed
Eckert and Strube-s (2000) es00 and Byron-s
(2002) phora. Furthermore Strube and M¨uller
(Strube and M¨uller, 2003) have presented a
machine learning approach for resolving APAs
in spoken dialogues.
1
Both es00 and phora
1
We do not discuss this approach further in this pa-
per.
recognise IPAsandAPAs on the basis of se-
mantic constraints on the argument position oc-
cupied by the anaphors and account for diﬀer-
ences in reference between personal and demon-
strative pronouns. In es00 demonstrative pro-
nouns preferentially refer to abstract entities,
while personal pronouns preferentially refer to
individual ones. es00 resolves IPAs applying
Strube-s (1998) algorithm. In phora the an-
tecedents of personal pronouns are searched for
looking at their degree of salience which is im-
plemented by word order as in (Grosz et al.,
1995). Demonstratives, instead, are searched
for in the list of activated entities (Gundel et al.,
1993) containing non NP antecedents, which are
assumed to be less salient. In phora demon-
stratives can also refer to Kinds.
es00 requires that the structure of dialogues
has been marked. Byron-s phora-algorithm
does not rely on predefined dialogue structure,
but only searches for abstract antecedents of
APAs in the sentence preceding the anaphor.
Thus it does not account for APAs referring to
larger discourse segments. phora relies on both
semantic knowledge and a model of speech acts
and accounts for more phenomena than es00.
Diﬀering from es00, phora has been imple-
mented.
To resolve IPAsandAPAsinDanishtexts
and dialogues Navarretta (2004) has proposed
the so-called dar-algorithm (Navarretta, 2004).
In dar APAs are resolved following the es00
strategy, but dar accounts for the Danish ab-
stract anaphors which occur in much more con-
texts than the English ones. Individual anaph-
ors are resolved in dar following a novel strat-
egy which combines models which identify high
degree of salience with high degree of givenness
(topicality) of entities in the hearer-s cognitive
model, e.g. (Grosz et al., 1995), with Hajiˇcov´a
et al.-s (1990) salience account which assigns the
highest degree of salience to entities in the focal
part of an utterance in Information Structure
terms. These entities often introduce new in-
formation in discourse.
In the present paper we describe an extended
version of the dar-algorithm accounting for dif-
ferences in the reference of various types of Dan-
ish demonstratives which we found analysing
the uses of pronouns in three text collections
(computer manuals, novels and newspaper ar-
ticles) and three corpora of recorded naturally-
occurring dialogues (the sl (Duncker and Her-
mann, 1996), the bysoc (Gregersen and Peder-
sen, 1991) and the pid corpora (Jensen, 1989)).
In the following we first discuss the background
for our proposal (section 2) then we describe
the extended dar-algorithm (section 3). In sec-
tion 4 we evaluate it and compare its perfor-
mance with that of other known algorithms. Fi-
nally we make some concluding remarks.
2 Background for DAR
In most applied approaches resolving pronom-
inal anaphors mainly consists in the following
steps: 1: determining the anaphoric antecedent
domain; 2: choosing the most prominent or
salient antecedent among possible candidates.
Thus determining the degree of salience of dis-
course elements, henceforth DEs, is essential to
anaphor resolution. Although there is not al-
ways an identity relation between linguistic an-
tecedents and referents, we also follow this strat-
egy, well aware that it is particularly problem-
atic for resolving APAs (se especially (Webber,
1991)). Nearly all salience-based algorithms
identify high degree of salience with high de-
gree of givenness of DEs. In fact, although dif-
ferent criteria are used for ranking DEs, such as
linear order, hierarchy of grammatical roles, in-
formation structure, Prince-s Familiarity Scale
(Prince, 1981), all algorithms assign the highest
prominence to the DEs which are most topi-
cal, known, bound, familiar and thus given,e.g.
(Grosz et al., 1995; Brennan et al., 1987; Strube,
1998). Analysing the Danish data we found a
restricted number of cases where high degree of
salience did not correspond to high degree of
topical ity, as it is the case in example (1). (1)
A: hvem...hvem arbejdede [din mor]
i
med?
(with whom... whom did [your mother]
i
work)
B: [Hun]
i
arbejdede med [vores nabo]
k
([She]
i
worked with [our neighbour]
k
)
[Hun]
k
var enke ... havde tre sønner [bysoc]
([She]
k
was a widow... had three sons) In (1)
the antecedent of the second occurrence of the
pronoun hun (she) is the object vores nabo (our
neighbour) which provides the information re-
quested in the preceding question. This nominal
is assigned lower prominence than the subject
pronoun hun (she)inmostsaliencemodels.
The only salience model which departs from
the givenness
2
assumption has been proposed
by Hajiˇcov´a et al. (1990). Hajiˇcov´a et al., in
fact, assign the highest degree of salience to DEs
in the focal part of an utterance in information
structure terms (Sgall et al., 1986). These enti-
ties often represent new information. Hajiˇcov´a
et al.-s approach is original and can account for
the data in (1), but it is problematic from an
applied point of view. In the first place it is
diﬃcult to determine the information structure
of all utterances. Secondly, focal candidate an-
tecedents are ranked highest in Hajiˇcov´a et al.-s
model, but they still compete with given can-
didate antecedents in their system. Finally the
data does not confirm that all entities in the fo-
cal part of an utterance have the highest degree
of accessibility.
We agree with Hajiˇcov´a-s insight, but in or-
der to operationalise the influence of focality in
a reliable way propose the following. Accessi-
bility by default is connected with givenness as
assumed in most algorithms. However, when
speakers explicitly change the degree of accessi-
bility of entities in discourse by marking them
as salient with information structure related de-
vices, these entities get the highest degree of
salience and are proposed as the preferred an-
tecedents of anaphors. In cases of explicit focus
marking the shift of focus of attention is as co-
herent as continuing speaking about the same
entities, because it is preannounced to the ad-
dressee. On the basis of the data we propose
a list of identifiable constructions in which ex-
plicit focus marking occurs.
3
Examples from
the list are the following:
a: Entities referred to by NPs which in Danish
are focally marked structurally (clefts, existen-
tial and topicalised constructions).
b: Entities referred to by NPs that follow fo-
cusing adverbs.
c: Entities focally marked by the prosody (if
this information is available) and/or entities
providing information requested in questions, as
in (1).
2
Giveness subsumes here concepts such as topicality
and familiarity.
3
Many of these constructions are also studied in the
Information Structure and in some anaphora resolution
literature, e.g. (Sidner, 1983).
Givenness preference in Danish can be mod-
elled by the hierarchy of verbal complements.
In addition to salience preferences we found
that parallelism can account for numerous uses
of Danish anaphors.
4
Inspired by the work of
(Kameyama, 1996) we have defined a prefer-
ence interaction model to be used in resolution.
Our model is given in figure 1.
5
The interac-
Parallel. ⊇ Focality ⊇ Pronom. chain⊇ Givenness
Figure 1: Interaction of preferences
tion model states that givenness preferences are
overridden by focality preference, when in con-
flict, and that they all are overridden by paral-
lelism.
dar also accounts for reference diﬀerences be-
tween Danish demonstrative and personal pro-
nouns. Weak (cliticised and unstressed) pro-
nouns usually refer to the most salient entity in
the utterance. Strong (stressed and demonstra-
tive) pronouns emphasise or put in contrast the
entities they refer to and/or indicate that their
antecedents are not the most expected ones.
6
Demonstratives preferentially refer to abstract
entities, while personal pronouns preferentially
refer to individual entities in ambiguous con-
texts. All these diﬀerences are also accounted
for in the literature on anaphora. However we
also found more language-specific peculiarities
in our data. Two examples of these pecular-
ities are the following. The Danish demon-
stratives denne/dette/disse (this common gen-
der/this neuter gender/these) never corefer with
a subject antecedent intrasententially. In the
few cases where they have a subject antecedent
in a preceding clause, there are no other an-
tecedent competitors. The abstract anaphor
dette, furthermore, is often used to refer to the
last mentioned situation in the previous sen-
tence, often expressed in a subordinated clause,
and not to the whole sentence or to an abstract
anaphor in the preceding sentence. The partic-
4
According to parallelism in adjacent utterances with
parallel grammatical complements, the preferred an-
tecedent of an anaphor in the second utterance is the
linguistic expression in the first utterance with the same
grammatical function.
5
Commonsense preferences which override all the
other preferences are not implemented.
6
The most frequent Danish third person singular gen-
der pronoun det can both be a personal pronoun (cor-
responding to it) and a demonstrative pronoun (corre-
sponding to this/that). In the latter case it is always
stressed.
ular phenomena are also accounted for in dar.
Approx. half of the APA occurrences in our
dialogues refer to entities evoked by larger dis-
course segments (more turn takings). Thus we
follow Eckert and Strube-s approach of mark-
ing the structure of dialogues and searching for
APA antecedents in the right frontier of the dis-
course tree (Webber, 1991). dar presupposes
diﬀerent discourse structures for texts and dia-
logues. dar follows the es00 and phora strat-
egy of discriminating between IPAsandAPAs
by rules looking at the semantic constraints on
the predication contexts in which the anaphors
occur. dar relies on more discriminating rules
than es00, which were defined on the basis of
large amounts of data and of the encodings of a
large computational lexicon.
dar uses language-specific rules to account
for Danish APAs. These occur in much more
contexts than in English where elliptical con-
structions or other anaphors such as too and so
are used. Examples of Danish-specific uses of
abstract anaphors are given in (2)-(3).
(2) Han var sulten. Det var jeg ikke. [pid]
(lit. He was hungry. That was I not)
(My friends were hungry. I wasn-t.)
(3) Han kunne svømme, men det kunne hun
ikke
(lit. He could swim, but it could she not)
(He could swim, but she couldn-t)
A language-specific rule recognising APAsis
the following: constructions with modal verbs
andanobject,suchasx skal man (lit. x shall
one) (one shall), x vil man (lit. x will one) (one
will).
3 The DAR Algorithm
3.1 Search Space and DE lists
dar presupposes the discourse structure de-
scribed by Grosz and Sidner (1986). The min-
imal discourse unit is the utterance U. Para-
graphs correspond to discourse segments in
texts. In dialogues discourse segments were
manually marked (se section 4). The dialogues
were structured with Synchronising Units (SU)
according to the definitions in ES00.
The immediate antecedent search space of a
pronoun x in utterance U
n
is the previous utter-
ance, U
n−1
.IfU
n
is the first component in SU
m
in dialogues the immediate search space for x is
SU
m−1
. dar assumes two antecedent domains
depending on whether the pronoun has or has
not been recognised as an IPA. The antecedent
domain for IPAsisfirstU
n−1
and then the pre-
ceding utterances in the right frontier of the dis-
course tree searched for in recency order.
7
The
antecedent domain for APAs or anaphors which
can both be IPAsandAPAsisU
n−1
.
dar operates on two lists of DEs, the Ilist
and the Alist.TheIlist contains the NPs re-
ferred to in U
n−1
ranked according to their de-
gree of salience and enriched with information
on gender, number, animacy and other sim-
ple semantic types necessary to implement se-
lectional restrictions. In the Ilist information
about the grammatical role of nominals is pro-
vided and strongly focally marked elements are
indicated. The leftmost element in the Ilist is
the most salient one. Givenness and focality
preferences are accounted for in the Ilist, as il-
lustrated in figure 2. Focally marked entities
are put in front of the list while the remaining
DEs are ordered according to verbal comple-
ment order. Inside verbal complements nomi-
nals are ordered according to their occurrence
order as illustrated in the second row of the fig-
ure. The abstract entities which are referred to
by an APA in U
n−1
or SU
m−1
are encoded in
the Alist. They are removed from the list af-
ter a new utterance (SU in dialogues) has been
processed if they have not been mentioned in it.
The context ranking for abstract entities is that
proposed by Eckert and Strube (2000).
3.2 The Algorithm and Its Functions
dar consists of two diﬀerent functions Re-
solveDet and ResolveIpa.Theformerisap-
plied if the actual pronoun x is third person
singular neuter, while the latter is applied in all
the remaining cases:
if x is singular & neuter
then go to ResolveDet(x)
else go to ResolveIpa(x)
The main steps of ResolveIpa are given in
figure 3. The ResolveIpa approach of indi-
cating possible reference ambiguities resembles
that proposed by Kameyama (1996). The main
structure of the function ResolveDet is in-
spired by es00. ResolveDet tests the pro-
noun x using the IPA and APA discriminating
rules discussed in section 2. ResolveDet is
simplified in figure 4. ResolveIpa-neu is like
ResolveIpa except that it returns if no NP an-
tecedents are found in U
n−1
(case A) so that
ResolveApa can be applied. ResolveApa
7
The search space in es00 is the preceding utterance
for all pronouns.
distinguishes between types of pronoun. If x
is weak, the preferred antecedent is searched
for among the elements indicated in the con-
text ranking, unless it is the object of the verb
gøre (do), modals, have (have) or the abstract
subject in copula constructions. In these cases
the pronoun is resolved to the VP of the ele-
ment in the A-list or in the context ranking.
If x is strong ResolveApa attempts to resolve
or classify it as vague depending on the type
of pronoun. This part of the algorithm is spe-
cific to Danish and accounts for the fact that
diﬀerent strong pronouns preferentially refer to
diﬀerent abstract entities in the data. Resolved
APAs are inserted into the Alist. In case of fail-
ure ResolveApa returns so that ResolveIpa-
neu can be applied. If both functions fail, the
pronoun is classified as vague.
3.3 Some Examples
In the following we look at the resolution of ex-
ample (1) from section 2. The simplified Ilists
and Alists after each utterance has been pro-
cessed are given in figure 5. (1) contains three
SUs. U
2
is an I/A thus it belongs to two syn-
chronising units (SU
1
and SU
2
). The Ilist after
U
1
has been processed, contains one element,
din mor (your mother). In U
2
the personal
pronoun hun (she) occurs, thus ResolveIpa is
applied. It resolves hun to the compatible NP
in the Ilist, din mor.AfterU
2
has been pro-
cessed the Ilist contains two elements in this
order: the focal marked entity vores nabo (our
neighbour) and the pronoun hun (= din mor).
ResolveIpa resolves the occurrence of the pro-
noun hun (she) in U
3
to the most salient can-
didate NP in the Ilist, vores nabo.Herefo-
cal preference overrides pronominal chain pref-
erence. Example (4) contains the APA det.
SU
1
: U
1
(I) U
2
(I/A):
U
1
: hvem...hvem arbejdede din mor med?
(with whom... whom did your mother work)
Ilist:[dinmor]
Alist:[]
—————————————————-
SU
2
: U
2
(I/A)
U
2
: Hun arbejdede med vores nabo
(She worked with our neighbour)
Ilist: [vores nabo,hun=din mor]
Alist:[]
—————————————————-
SU
3
: U
3
(I)
U
3
: Hun var enke ... havde tre sønner
(She was a widow... had three sons)
Ilist: [hun=vores nabo,tre sønner]
Alist:[]
Figure 5: IlistsandAlists for example (1)
(4): Du har svært ved at se musemarkøren p˚a
skærmen. Hvordan klarer du det? [edb]
(You have diﬃculties seing the mouse-cursor
FOCAL MARKED > SUBJECT > OBJECT/PrepOBJECT > OBJECT 2 > OTHER COMPLS > ADJUNCTS
————————————————————————————————————————————————-
NCOMPL
1
>
prec
NCOMPL
2
>
prec
...>
prec
NCOMPL
n
Figure 2: Order of DEsintheIlist
antecedent in Ilist
for U
n−1
or S
m−1
?
no antecedent 1 anteced. more anteced.
look in prec.
Ilists apply preferences
(1) parallelism?
ant. found ant. not found return it yes no
return it x =inferable return y =retury =
parallel leftmost
candidate and cand. in Ilist
candid. list and c. list
x is weak?
yes no
return y return second
and candidate and
list ambig. list ambig.
Figure 3: ResolveIpa
(common-gend) on the screen (common-gend).
How do you manage it/this (neuter gender))?
The simplified IlistsandAlistsafterthetwo
utterances in (4) have been processed are pre-
sented in figure 6. After U
1
has been pro-
U
1
: Du har svært ved at se musemarkøren p˚askærmen.
Ilist: [musemarkøren, skærmen]
Alist:[]
—————————————————————-
U
2
: Hvordan klarer du det?
Ilist:[]
Alist:[det=U
1
]
Figure 6: IlistsandAlists for example (4)
cessed there are two common gender singular
NPs in the Ilist, musemarkøren (the mouse
cursor) and skærmen (the screen). In U
2
the
singular neuter gender pronoun det (it) occurs,
thus ResolveDet is applied. The pronoun
is neither IPA nor APA according to the dis-
criminating rules. ResolveDet attempts to
find an individual antecedent of the weak pro-
noun, applying the function ResolveIpa-neu.
ResolveIpa-neu fails because the two DEsin
the Ilist do not agree with the pronoun. Then
the function ResolveApa resolves x looking at
the context ranking. Being the Alist empty, U
1
,
is proposed as antecedent. The resolved APA
is added to the Alist.
4 Tests and Evaluation
We have manually tested dar on randomly cho-
sen texts and dialogues from our collections.
The performance of dar on dialogues has been
compared with that of es00. The function
for resolving IPAs(ResolveIpa) has similarly
been tested on texts, where APAswereex-
cluded. We have compared the obtained re-
sults with those obtained by testing bfp (Bren-
nan et al., 1987) and str98 (Strube, 1998).
In all tests the intrasentential anaphors have
been manually resolved and expletive and cat-
aphoric uses of pronouns have been marked and
excluded from the test. Dialogue act units were
marked and classified by three annotators fol-
lowing (Eckert and Strube, 2000). The relia-
bility for the two annotation tasks (κ-statistics
(Carletta, 1996)) was of 0.94 and 0.90 respec-
tively. Pronominal anaphors were marked, clas-
sified and resolved by two annotators. The κ-
statistics for the pronoun classification was 0.86.
In few cases (one in the texts and two in the dia-
logues) where the annotators did not agree upon
resolution, the pronouns were marked as am-
biguous and were excluded from the test. The
results obtained for bfp and str98 are given in
table 1, while the results of dar-s ResolveIpa
are given in table 2. In the tables CR stands for
“correctly resolved”, HR stands for “resolved
by humans”, RA stands for “resolved over all”,
P stands for precision and R stands for recall.
Because dar both classifies and resolves anaph-
ors, both precision and recall (respect to hu-
man resolution) are given in table 2. The re-
sults indicate that ResolveIpa performs sig-
nificantly better than bfp and str98 on the
type of x?
IPA APA IPA/APA?
ResolveIPA-new ResolveAPA x weak?
x weak? yes no
yes no ResolveIpa-new ResolveApa
look in Alist x = det x = dette x = det her
resolve look in Alist look in Alist mark
as es00 as es00 prefer subclause as vague
Figure 4: ResolveDet
Danish texts. The better performance of dar
was due to the account of focal and parallelism
preferences, of the diﬀerent reference mecha-
nisms of personal and demonstrative pronouns
and to the enlarged resolution scope. Further-
more dar recognises some generic pronouns and
inferable pronouns and excludes them from res-
olution, but often fails to recognise antecedent-
less and inferable plural pronouns, because it
finds a plural nominal in the preceding discourse
and proposes it as antecedent. The lack of com-
monsense knowledge explains many incorrectly
resolved anaphors. The results of the test of
algorithm CR HR P
bfp 513 645 79.53
str98 524 645 81.24
Table 1: Results of bfp and str98 on texts
CR RA HR P R
575 651 645 88.33 89.14
Table 2: Results of ResolveIpa on texts
the dar algorithm on written texts are in ta-
ble 3. These results are good compared with the
results of the function ResolveIpa (table 2).
The discriminating rules identify correctly IPAs
and APAs in the large majority of the cases.
Recognition failure often involves pronouns in
contexts which are not covered by the discrim-
inating rules. In particular dar fails to resolve
singular neuter gender pronouns with distant
antecedents and to identify vague anaphors, be-
cause it always “finds” an antecedent in the con-
text ranking. Correct resolution in these cases
requires a deep analysis of the context. The
results of applying dar and es00 on Danish di-
alogues are reported in table 4.
8
The results of
the tests indicate that dar resolves IPAssig-
nificantly better than es00 (which uses str98).
8
We extended es00 with the Danish-specific identifi-
cation rules before applying it.
resolution IPA
CR RA HR P R
560 651 645 86.02 86.82
resolution APA
CR RA HR P R
68 87 77 78.16 88.31
Table 3: Results of dar on texts
dar correctly resolves more Danish demonstra-
tive pronouns than es00, because it accounts
for language-specific particularities. In general,
however, the resolution results for APAsare
similar to those obtained for es00.Thisisnot
surprising, because dar usesthesameresolu-
tion strategy on these pronouns. dar performs
better on texts than on dialogues. This reflects
the more complex nature of dialogues. The re-
sults indicate that the IPA/APA discriminat-
ing rules also work well on dialogues. The cases
of resolution failure were the same as for the
texts. As an experiment we applied dar on the
resolution IPA
Algorithm CR RA HR P R
ES00 258 411 414 62.77 62.31
DAR 289 386 414 74.87 68.81
resolution APA
Algorithm CR RA HR P R
ES00 179 286 269 62.59 66.54
DAR 199 277 269 71.84 73.98
Table 4: Results of es00 and dar on dialogues
dialogues without relying on the predefined di-
alogue structure. In this test the recognition of
IPAsandAPAs was still good, however the suc-
cess rate for IPAs was of 60.1 % and for APAs
was of only 39.3%. Many errors were due to
the fact that antecedents were searched for in
the preceding discourse in linear order and that
ungrounded utterances were included in the dis-
course model.
5 Concluding Remarks
In this paper we presented dar, an algorithm
for resolving IPAsandAPAsinDanishtexts
and dialogues. In dar diﬀerences between the
referential characteristics of Danish weak and
strong pronouns are accounted for and a novel
strategy for resolving individual anaphors is
proposed. This strategy combines givenness
with focality preferences to model salience and
also accounts for parallelism preferences. dar
performs significantly better on IPAsthanal-
gorithms which only rely on givenness-based
salience models.
dar extends the es00 strategy of classify-
ing and resolving (some types of) APAs. The
tests of dar indicate that the es00-s approach
of recognising APAsisalsopromisingfortexts
and other languages than English.
dar has not been compared with phora
which is the only abstract anaphora algorithm
implemented. We find the algorithm very inter-
esting because it addresses many of the same
phenomena, but with diﬀerent strategies. It
would be useful to combine some of these strate-
gies with the approaches proposed in dar and
es00 to improve the still problematic resolution
of abstract anaphors.

References

S. F. Brennan, M. W. Friedman, and C. J. Pollard. 1987. A Centering Approach to Pronouns. In Proceedings of the ACL-87, pages
155-162, CA.

D. Byron and J. Allen. 1998. Resolving demonstrative pronouns in the trains93 corpus. In
Proceedings of DAARC 2, pages 68-81.

D. K. Byron. 2002. Resolving Pronominal Reference to Abstract Entities. In Proceedings of
the ACL 2002.

J. Carletta. 1996. Assessing agreement on classification tasks. The kappa statistic. In Computational Linguistics, 22(2):249-254.

D. Duncker and J. Hermann. 1996. Patien-
tord og lgeord - srord eller fllesord?
Manedsskrift for Praktisk Lgegerning, pages
1019-1030.

M. Eckert and M. Strube. 2000. Dialogue acts,
synchronising units and anaphora resolution.
Journal of Semantics, 17:51-89.

F. Gregersen and I. L. Pedersen, editors. 1991.
The Copenhagen study in urban sociolinguis-
tics. Reitzel.

B. J. Grosz and C. L. Sidner. 1986. Attention,
Intentions, and the Structure of Discourse.
Computational Linguistics, 12(3):175-284.

B. Grosz, A. K. Joshi, and S. Weinstein. 1995.
Centering:A Framework for Modeling the Local Coherence of Discourse. Computational
Linguistics, 21(2):203-225.

J. K. Gundel, N. Hedberg, and R. Zacharski.
1993. Cognitive status and the form of referring expressions in discourse. Language,
69(2):274-307.

E. Hajicova, P. Kubon, and V. Kubon. 1990.
Hierarchy of Salience and Discourse Analysis and Production. In H. Karlgren, editor,
Proceedings of COLING-90, volume III, pages
144-148, Helsinki.

K. A. Jensen. 1989. Projekt invandrerdansk.
Technical report, Copenhagen University.

M. Kameyama. 1996. Indefeasible Semantics
and Defeasible Pragmatics. In M. Kanazawa,
C. Pinon, and H. de Stwart, editors, Quanti-
fiers, Deduction and Context, pages 111-138.
CSLI, Stanford, CA.

C. Navarretta. 2002. Combining Information Structure and Centering-based Models of Salience for Resolving Intersentential Pronominal Anaphora. In A. Branco,
T. McEnery, and R.Mitkov, editors, Proceedings of the 4th DAARC, pages 135-140.
Edicoes Colibri.

C. Navarretta. 2004. . In Proceedings of Reference Resolution and Its Applications Workshop at ACL 2004, Barcelona, Spain.

E. F. Prince. 1981. Toward a taxonomy of
given-new information. In P. Cole, editor,
Radical Pragmatics, Academic Press, pages
223-255.

C. Sidner. 1983. Focusing in the Comprehen-
sion of Definite Anaphora. In M. Brady and
R. Berwick, editors, Computational Models of
Discourse, MIT Press, pages 267-330.

P. Sgall, E. Hajicova, and J. Panevova. 1986.
The Meaning of the Sentence in its Semantic
and Pragmatic Aspects. Reidel, Dordrecht.

M. Strube. 1998. Never Look Back: An Al-
ternative to Centering. In Proceedings of
COLING-ACL-98, II, pages 1251-1257.

M. Strube and C. Muller. 2003. A machine learning approach to pronoun resolu-
tion in spoken dialogue. In Proceedings of the
ACL-03, pages 168-175.

B. L. Webber. 1991. Structure and Ostension in the Interpretation of Discourse Deixis.
Natural Language and Cognitive Processes,
6(2):107-135, January.
