Ontology-based linguistic annotation
Philipp Cimiano, Siegfried Handschuh
Institute AIFB
University of Karlsruhe
a0 cimiano,handschuh
a1 @aifb.uni-karlsruhe.de
Abstract
We propose an ontology-based framework
for linguistic annotation of written texts.
We argue that linguistic annotation can be
actually considered a special case of se-
mantic annotation with regard to an on-
tology such as pursued within the con-
text of the Semantic Web. Furthermore,
we present CREAM, a semantic annota-
tion framework, as well as its concrete im-
plementation OntoMat and show how they
can be used for the purpose of linguistic
annotation. We demonstrate the value of
our framework by applying it to the an-
notation of anaphoric relations in written
texts.
1 Introduction
Linguistic annotation is crucial for the develop-
ment and evaluation of natural language processing
(NLP) tools. In particular machine-learning based
approaches to part-of-speech tagging, word sense
disambiguation, information extraction or anaphora
resolution - just to name a few - rely on corpora an-
notated with the corresponding phenomenon to be
trained and tested on. In this paper, we argue that
linguistic annotation can to some extent be consid-
ered a special case of semantic annotation with re-
gard to an ontology. Part-of-Speech (POS) annota-
tion for example can be seen as the task of choos-
ing the appropriate tag for a word from an ontol-
ogy of word categories (compare for example the
Penn Treebank POS tagset as described in (Marcus
et al., 1993)). The annotation of word senses such
as used by machine-learning based word sense dis-
ambiguation (WSD) tools corresponds to the task of
selecting the correct semantic class or concept for a
word from an underlying ontology such as WordNet
(Resnik, 1997). Annotation by template filling such
as used to train machine-learning based information
extraction (IE) systems as (Ciravegna, 2001) can be
seen as the task of finding and marking all the at-
tributes of a given ontological concept in a text. An
ontological concept in this sense can be a launching
event, a management succession event or a person
together with attributes such as name, affiliation, po-
sition, etc. The annotation of anaphoric or bridging
relations is actually the task of identifying the se-
mantic relation between two linguistic expressions
representing a certain ontological concept.
Most linguistic annotation tools make use of schema
specifying what can actually be annotated. These
schema can in fact be understood as a formal rep-
resentation of the conceptualization underlying the
annotation task. Ontologies are formal specifica-
tions of a conceptualization (Gruber, 1993) so that
it seems straightforward to formalize annotation
schemes as ontologies and make use of semantic an-
notation tools such as OntoMat (Handschuh et al.,
2001) for the purpose of linguistic annotation.
The structure of this paper is as follows: Section 2
presents the ontology-based framework for linguis-
tic annotation, and section 3 shows how the frame-
work can be applied to the annotation of anaphoric
relations. Section 4 presents CREAM, a semantic
annotation framework for the Semantic Web as well
as its concrete implementation OntoMat. Finally,
section 5 discusses related work, and section 6 con-
cludes the paper.
2 The Ontology-based linguistic
annotation framework
An ontology is a formal specification of a conceptu-
alization (Gruber, 1993). A conceptualization can be
understood as an abstract representation of the world
or domain we want to model for a certain purpose.
The ontological model underlying this work is basi-
cally the one in (Bozsak et al., 2002). According to
this model, an ontology is defined as follows:
Definition 1 (Ontology)
An ontology is a structure a0a2a1a4a3 a5a7a6a9a8a11a10a13a12a14a8a16a15a17a8a11a10a19a18a21a20
consisting of (i) two disjoint sets a6 and a15 called
concept identifiers and relation identifiers respec-
tively, (ii) a partial order a10a9a12 on a6 called concept
hierarchy or taxonomy, (iii) a function a22a23a1a24a15a26a25
a6a28a27a29a6
1 called signature and (iv) a partial order
a10a30a18
on a15 called relation hierarchy.
In addition, the underlying ontological model also
allows to define axioms:
Definition 2 (a31 -Axiom System)
Let a31 be a logical language. An a31 -axiom system
for an ontology a0 as defined above is a pair (AI,a32 )
where (i) AI is a set whose elements are called axiom
identifiers and (ii) a32 : AI a25 a31 is a mapping. The
elements of A:=a32 (AI) are called axioms.
In our ontological framework, a relation a33 can for ex-
ample be defined as symmetric, i.e. SYM(a33 ). Now
if F-logic (Kifer et al., 1995) is used as the underly-
ing logical language such as in (Staab and M¨adche,
2000), the translation of the SYM axiom identifier is
as follows:
a34
a15a17a8a36a35a37a8a16a38a39a38a41a40a15a43a42a44a35a46a45a48a47a50a49a51a38a53a52a39a5a54a15a55a20 (1)
a56a58a57a60a59
a35a61a40a15a43a42a44a38a55a45
In addition, we will also distinguish special type of
relations which we will call attributes. These are re-
lations with a plain datatype as range, i.e. relations
a56a37a62a29a63 with signatures of the type
a22a64a1
a63
a25a65a6a66a27a68a67 ,
1Here we actually restrict the model to binary relations.
where a67 is a plain datatype such as a string, an inte-
ger, etc.
Our framework basically offers three ways of anno-
tating a text with regard to an ontology:
a69 a linguistic expression appearing in a text can
be annotated as an instance of a certain onto-
logical concept a70 a62 a6
a69 a linguistic expression in a text can be anno-
tated as an attribute instance of some other
linguistic expression previously annotated as a
certain concept a70 a62 a6
a69 the semantic relation between two linguistic ex-
pressions respectively annotated as instances of
two concepts a70a72a71a11a8a16a70a74a73 a62 a6 can be annotated as an
instance of relation a33 a62 a15 if a22a75a5a76a33a77a20a51a3a28a5a54a70a78a71a79a8a16a70a74a73a79a20
The advantages of an ontology-based linguistic an-
notation framework as described above are the fol-
lowing:
a69 The formalization of the annotation scheme
as an ontology as well as the use of standard
formalisms such as RDF (Lassila and Swick,
1999) or OWL 2 to encode it, allow to reuse the
scheme across different annotation tools. This
meets the interoperability requirement men-
tioned in (Ide, 2002).
a69 The specification of the annotation task, i.e. the
annotation scheme, can be performed in an ar-
bitrary ontology development environment and
thus becomes completely independent of the
annotation tool actually used.
a69 The ontology-based linguistic annotation
model offers the kind of flexibility mentioned
in (Ide, 2002) in the sense that it is general
enough to be applied to a broad variety of
annotation tasks.
a69 The fact that annotation is performed with re-
spect to an ontological hierarchy offers anno-
tators the possibility to choose the appropri-
ate level of annotation detail such that they are
never forced to overspecify, i.e. to annotate
more specifically than they actually feel com-
fortable with.
2http://www.w3.org/TR/owl-ref/
In addition, a hierarchical annotation offers further
possibilities regarding the computation of the agree-
ment between different annotators as well as the
evaluation of a system against a certain annota-
tion. In this sense, instead of measuring only the
categorial agreement between annotators with the
kappa statistic (Carletta, 1996) or the performance
of a system in terms of precision/recall, we could
take into account the hierarchical organization of the
categories or concepts by making use of measures
considering the ’hierarchical distance’ between two
concepts such as proposed by (Hahn and Schnat-
tinger, 1998) or (M¨adche et al., 2002).
Furthermore, the use of an ontology-based and thus
more semantic framework for linguistic annotation
has two further, very interesting properties. On the
one hand, the use of an ontology helps to constrain
the possible relations between two concepts, thus re-
ducing the amount of errors in the annotation pro-
cess. For example when annotating Coreference-
relations in a text, it seems obvious that an event
and an entity will never be coreferring and in fact
such an erroneous annotation can be avoided if the
underlying ontological model actually forbids such
an annotation (see below). Furthermore, by using
axioms such as described above for example stating
that Coreference is reflexive, symmetric and tran-
sitive - thus representing an equivalence relation -
the evaluation of systems becomes much easier and
more straightforward when using an inference ma-
chine such as (Decker et al., 1999). If an annotator
for example annotates the following coreferences:
Coreference(A,B) and Coreference(B,C) a system’s
answer such as Coreference(A,C) will actually be
counted as correct due to the fact that Coreference is
defined as a transitive relation within the ontology.
3 Annotating anaphoric relations
Before showing how our framework can be applied
to the annotation of anaphoric relations in written
texts, the assumptions underlying our model have
to be explained. First, we aim at a more semantic
annotation of anaphoric relations than for example
described in (M¨uller and Strube, 2001) because we
think that such a model can to some extent be sub-
sumed by the one we propose. In fact, we will un-
derstand the term anaphoric in a much wider sense
in line with (Krahmer and Piwek, 2000) and (van
Deemter and Kibble, 2000). They argue for exam-
ple that coreference is not a necessary property of
anaphora such as proposed in (M¨uller and Strube,
2001). So annotating the relation between two ex-
pressions as anaphoric will correspond to the most
general relation in our hierarchy. In particular, in our
model Identity or Coreference will only be a special
type of anaphoric relation (compare figure 2).
On the other hand, bridging will be defined in our
framework in line with (Asher and Lascarides, 1999)
as “the inference that two objects or events that
are introduced in a text are related in a particular
way that isn’t explicitly stated”. Thus Coreference
or Identity can represent an anaphoric relation or
more specifically a bridging reference depending on
whether the identity relation is explicit or not. Con-
sider the following minimal pair:
(2) John bought a car yesterday. The car was in a
good state.
(3) John bought a car yesterday. The vehicle was
in a good state.
In example (2), the anaphoric relation is explicit
due to the matching heads of the NPs a car and
The car. In (3) the anaphoric or bridging rela-
tion is not explicit as world knowledge such as that
cars are vehicles is needed to resolve the reference.
In the semantics-based model for the annotation of
anaphoric relations we propose in this paper, both
examples will in fact be annotated as instances of
the Coreference or Identity relation. Consequently,
we will completely omit the concept bridging refer-
ence in the ontology underlying the annotation. In
fact, we claim that the classification of an anaphora
as a bridging reference, direct anaphora, pronomi-
nal anaphora, etc. such as pursued in (M¨uller and
Strube, 2001) can be seen as a byproduct of a more
semantic classification as proposed here if additional
grammatical information provided by the annotators
is available. This grammatical information can be
added to the concepts depicted in figure 2 in form
of attributes specifying the grammatical form of the
expression, i.e. whether it is for example a noun,
an NP, a pronoun, a verb or a VP, as well as in-
formation about its head, gender or tense. The se-
mantic classification proposed here together with the
grammatical information modeled as attributes of a
concept will then yield a classification as envisioned
by (M¨uller and Strube, 2001). For example, if two
expressions are annotated as coreferring, this se-
mantic relation can be further classified as nominal
anaphora if the referring expression is a pronoun,
as direct anaphora if the heads of the expression
match or as a bridging reference otherwise. On the
other hand, all the Non-Identity relations modeled
in the ontology underlying the annotation task will
lead to a classification as a bridging reference (com-
pare figure 2). However, it should be mentioned that
we do not aim at such a ’grammatical’ classifica-
tion of anaphoric relations. We envision a task as in
(Asher and Lascarides, 1999), where bridging ref-
erence resolution corresponds to the task of finding
the discourse referent serving as antecedent as well
as the semantic relation between this discourse ref-
erent and the one of the referring expression.
In our model, an expression can be antecedent for
more than one referring expression, an assumption
which seems to be commonly shared by many anno-
tation schema. However, in our model a certain ex-
pression can also refer to more than one antecedent.
(Poesio and Reyle, 2001) for instance show that the
antecedent of a referring expression can in fact be
ambiguous in a way that the overall interpretation
of the expression or sentence is not affected. Fur-
thermore, (Poesio and Reyle, 2001) argue that it is
not clear whether the addressees of an utterance ac-
tually are aware of all the possible antecedents for a
certain referring expression, if they underspecify the
antecedent of a referring expression in case the over-
all interpretation is not affected or if they just choose
one of the possible antecedents without being aware
of the other ones. In any case, a model for the anno-
tation of anaphoric or bridging relations should not
a priori exclude that referring expressions can have
more than one antecedent. Consequently, the anno-
tation of the semantic relation between a referring
expression and an antecedent can neither take place
at the antecedent nor the referring expression such
as in (M¨uller and Strube, 2001), but in a functional
way, i.e. at a virtual edge between them.
The ontology underlying our annotation scheme
is depicted schematically in figure 1 We distinguish
two types of eventualities: events and states, and
model the discourse relations described in (Las-
entities
top
part_of
events states
eventualities
entity  intensionalentitysets of entities
member_of
value_of
role_of
explanation,
elaboration,
narration
background
result
Figure 1: The ontology underlying the annotation
scheme
carides and Asher, 1991) as semantic relations be-
tween them. In addition, we distinguish between
three types of (meta-) entities: sets of entities, inten-
sional entities (van Deemter and Kibble, 2000) and
(real-world) entities together with the potential rela-
tions such as member of, part of, etc. between them
as well as to other types: An entity for example can
play a certain thematic role in some event (compare
figure 1).
With such a concept hierarchy as well as seman-
tic relations with a precisely defined signature, we
can for example overcome annotation problems of
intensionality and predication as discussed in (van
Deemter and Kibble, 2000). In order to profit from
the benefits of a hierarchical annotation, we also de-
fine a hierarchy on the semantic relations (see figure
2). Thus if annotators for example feel that there
is an anaphoric relation between two linguistic ex-
pressions, but can not specify the type of relation,
they can choose the most general relation in the hi-
erarchy, i.e. anaphoric relation. As mentioned in
section 2, the idea is that annotators are never forced
to overspecify and can annotate at the hierarchical
level they feel comfortable with.
4 CREAM and OntoMat
CREAM is an annotation and authoring framework
and OntoMat-Annotizer (OntoMat for short) is its
concrete implementation. The framework itself was
developed for the creation of ontology-based anno-
tation in the context of the Semantic Web. Its main
objective is thus the transformation of existing syn-
relations
rhethorical
Coreference/
Identitiy
anaphoric
relations
Non−Identity
value
resultnarration background explanation elaboration
member part_of role
Figure 2: The hierarchical organization of the se-
mantic relations.
tactic resources (viz. textual documents) into inter-
linked knowledge structures that represent relevant
underlying information (Handschuh et al., 2001).
However, with an apropriate ontology one can also
take advantage of the framework and use it for lin-
guistic annotation. In the subsequent section we will
explain only the features that are relevant to this pur-
pose.
4.1 CREAM Features
4.1.1 User Interface
OntoMat’s document viewer visualizes the docu-
ment contents. The user may easily provide annota-
tions by selecting pieces of text and aligning it with
parts of the ontology. The document viewer supports
various formats3 (HTML, PDF, XML, etc.). The
Ontology and Fact Browser is the visual interface
to the ontology and the annotated facts. The anno-
tation framework needs guidance from the ontology.
In order to allow for sharing of knowledge, newly
created annotations must be consistent with a given
ontology. Otherwise, if annotators instantiate arbi-
trary classes and properties the semantics of these
properties remains void and the annotation thus use-
les.
Both the Ontology and Fact Browser and the docu-
ment editor/viewer are intuitive to use: Drag’n’drop
helps to avoid syntax errors and typos and a good
visualization of the ontology helps the annotators to
correctly choose the most appropriate class for an
3The current OntoMat implementation is restricted to
HTML/XHTML and plain text. A support for PDF is in de-
velopment.
instance (compare figure 3).
4.1.2 Annotation
An annotation in our context is a set of instantia-
tions of classes, relationships and attributes. This in-
stances are not directly embedded into the text, but
are pointing to appropriate fragments of the docu-
ment. The link between the annotation and the doc-
ument is done by using XPointer (DeRose et al.,
2001) as a adressing mechanism. This has some ad-
vantages with regards to the flexibility of annotation
as it allows (i) multiple annotation (ii) nested anno-
tation and (iii) overlapping annotation of text seg-
ments.
4.1.3 Annotation Inference Server
The annotation inference server reasons on the in-
stances and on the ontology. Thereby, it also takes
into account the axioms modeled within the ontol-
ogy and can thus be used in the evaluation of a
certain system such as described in section 2. We
use Ontobroker’s F-Logic-based inference engine
(Decker et al., 1999) as annotation inference server.
The F-Logic inference engine combines ordering-
independent reasoning in a high-level logical lan-
guage with a well-founded semantics.
4.1.4 Storage
CREAM supports different ways of storing the
annotation. This flexiblity is given by the XPointer
technique which allows to separate the annotation
from the document. Hence, the annotations can be
stored together with the document. Alternatively
or simultaneously it is also possible to store them
remote, either in a separate file or in the annotation
inference server.
4.2 Annotaing anaphoric relations
The ontology described in section 3 is available in
the form of DAML+OIL4 classes and properties, in
OWL, as pure RDF-Schema and in F-Logic. In the
following, we shortly explain how OntoMat can be
used for the creation of instances consistent with the
ontology described in section 3.
Figure 3 shows the screen for navigating the ontol-
ogy and creating annotations in Ontomat. The right
4http://annotation.semanticweb.org/ontologies/AnaphOnto.daml
Figure 3: Annotation Tool Screenshot.
pane displays the document and the left panes shows
the ontological structures contained in the ontology,
namely classes, attributes and relations. In addition,
the left pane shows the current semantic annotation
knowledge base, i.e. existing class instances, at-
tribute instances and relationship instances created
during the semantic annotation. First of all, the user
browses a document by entering the URL of the web
document that he would like to annotate. Then he
loads the corresponding ontology into the ontology
browser. He selects a text fragment by highlighting
it. There are two possibilities for the text fragment
to be annotated: as an instance or as a relation. In
the case of an instance, the user selects in the on-
tology the class where the text fragment fits in, e.g.
for the expression ”a car” in example 2, he would
select the class entity. By clicking on the class, the
annotation gets created and thus the text fragment
will be displayed as an instance of the selected class
in the ontology browser. The relationships between
the created instances can be specified, e.g. the en-
tity The car can be annotated as coreferring with the
preceding entity a car as described in section 2. For
this purpose, when selecting a certain class instance
as well as a corresponding semantic relation from
the ontology, OntoMat already presents the possible
target class instances according to the range restric-
tions of the chosen relation. Hereby erroneous an-
notations of relations are avoided (compare section
2). Futhermore, literal attributes can be assigned to
every created instance by typing them into the re-
lated attribute field. The choice of the predefined
attributes depends on the class the instance belongs
to. Thereby, instances of a certain concept can be
annotated with grammatical information about how
they are linguistically expressed, i.e. through an NP,
a noun, a pronoun, a verb, etc. (compare section 3).
5 Discussion of Related Work
There is a vast amount of frameworks and tools
developed for the purpose of linguistic annotation.
However, in this paper we will focus on the discus-
sion of frameworks for the annotation of anaphoric
or discourse relations in written texts. In the an-
notation scheme proposed by (M¨uller and Strube,
2001) in the context of their annotation tool MMAX
and in contrast to the one proposed in this paper,
anaphoric relations are restricted to coreferring ex-
pressions, while bridging relations are restricted to
non-coreferring ones. In line with (Krahmer and Pi-
wek, 2000) and (van Deemter and Kibble, 2000) this
is in our view a too strict definition of anaphora so
that we propose a more relation-based classification
of anaphoric and bridging relations. Furthermore, in
(M¨uller and Strube, 2001), anaphoric relations are
further differentiated according to the lexical items
taking part in the relation. We have shown that un-
der the assumption that the corresponding grammat-
ical information is provided by the annotators, such
a classification can be seen as a byproduct of a more
semantic one such as outlined in this paper. In ad-
dition, (M¨uller and Strube, 2001) propose to spec-
ify antecedence with regard to equivalence classes
rather than with regard to particular antecedents.
However, this has the disadvantage that the infor-
mation about the actual antecedent an annotator has
selected is actually lost. Thus in our annotation pro-
posal the fact that the Coreference relation forms
equivalence classes is modeled by an underlying ax-
iom system which can be exploited in the evaluation
of a system against the annotation standard.
The annotation scheme proposed by Poesio et al.
(Poesio and Vieira, 1998) is a product of a corpus-
based analysis of definite description (DD) use
showing that more than 50% of the DDs in their cor-
pus are discourse new or unfamiliar. Thus in Poesio
et al.’s annotation scheme definite descriptions are
also explicitly annotated as discourse new.
The MUC coreference scheme (Hirschman and
Chinchor, 1997) is restricted to the annotation of
coreference relations, where coreference is also de-
fined as an equivalence relation. Though this anno-
tation scheme may seem quite simple, we agree with
(Hirschman and Chinchor, 1997) that it is complex
enough when taking into account the agreement of
the annotators on a task. In fact, it has been shown
that the agreement of subjects annotating bridging
(Poesio and Vieira, 1998) or discourse (Cimiano,
2003) relations can be too low for tentative conclu-
sion to be drawn (Carletta, 1996). The motivation of
the MUC coreference scheme was thus to develop
an annotation scheme leading to a good agreement.
On the other hand, our motivation is to show how
our ontology-based framework can be applied to the
annotation of anaphoric relations in written texts and
from this perspective the MUC coreference annota-
tion scheme would have been in fact too restricted to
actually show all the advantages of our approach.
The UCREL (Fligelstone, 1992) and DRAMA (Pas-
soneau, 1996) annotation schemes are more related
to ours than the schemes above in the sense that they
also provide a rich set of particular bridging rela-
tions that can be annotated. However, in contrast to
the ontology-based framework presented in this pa-
per, these bridging relations are not constrained with
regard to the conceptual types of their arguments, so
that erroneous annotations can not be avoided.
The coreference annotation scheme proposed within
the MATE Workbench project consists of a core as
well as an extended scheme (Davies et al., 1998).
The core scheme is in principle identical with the
MUC coreference scheme and is restricted to the an-
notation of coreference in the sense of (van Deemter
and Kibble, 2000). The extended scheme also al-
lows the annotation of bound anaphors, of the rela-
tionship between a function and its values, of differ-
ent set, part and possession relations, of instantiation
relations as well as of event relations. The MATE
scheme is related to our ontology-based annotation
scheme in the sense that relations are also annotated
as triples via the link-tag (Davies et al., 1998). As
in our framework, the MATE scheme also allows
to mark up ambiguities of reference. However, in
contrast to the MATE scheme our framework has no
means to specify a preference order on these am-
biguous antecedents. On the other hand, the MATE
scheme also includes a reasonable and complete tax-
onomy of markables as well as some features rele-
vant for the annotation of coreference in dialogues
such as the treatment of hesitations, disfluencies and
misunderstandings.
6 Conclusion
We have argued that many linguistic annotation
tasks can be seen as a special case of semantic an-
notation with regard to an ontology and have pro-
posed a novel ontology-based framework for this
purpose. We have furthermore applied our frame-
work to the annotation of anaphoric relations in writ-
ten texts. For this purpose we have proposed a rela-
tively complex annotation scheme for anaphoric re-
lations in which we have deliberatively abstracted
from important issues such as inter-annotator agree-
ment. In fact, the main contribution of this paper is
certainly not the annotation scheme proposed in sec-
tion 2, but to show that relatively complex annota-
tion schemes such as the one proposed can be mod-
eled in our ontology-based framework in a straight-
forward manner. The main benefits of the approach
presented here are that the annotation can be per-
formed at different levels of detail with regard to
a given taxonomy as well as that the possible rela-
tions between two different concepts are constrained
by the underlying ontology, which could make the
annotation less error-prone. Furthermore, we have
shown how the modeling of axioms within the on-
tology can actually make the evaluation of a system
more straightforward. The most important advan-
tage is that by specifying the annotation scheme in
form of an ontology and adhering to standards such
as RDF or OWL, it can be easily exchanged between
different parties and can also be developed inde-
pendently of the annotation tool used, which meets
the interoperability requirement mentioned in (Ide,
2002). In addition, our framework is flexible enough
to be applied to various annotation tasks, which is
also a requirement mentioned in (Ide, 2002). In the
future, we hope to show that, with the necessary ex-
tensions, our model is also suitable for the annota-
tion of multi-modal corpora as well as of speech sig-
nals.

References
N. Asher and A. Lascarides. 1999. Bridging. Journal of
Semantics, 15:83–113.
E. Bozsak, M. Ehrig, S. Handschuh, A. Hotho,
A. M¨adche, B. Motik, D. Oberle, C. Schmitz, S. Staab,
L. Stojanovic, N. Stojanovic, R. Studer, G. Stumme,
Y. Sure, J. Tane, R. Volz, and V. Zacharias. 2002.
Kaon - towards a large scale semantic web. In Pro-
ceedings of the Third International Conference on
E-Commerce and Web Technologies (EC-Web 2002).
Springer Lecture Notes in Computer Science.
J. Carletta. 1996. Asessing agreement on classification
tasks: The kappa statistic. Computational Linguistics,
22(2):249–254.
P. Cimiano. 2003. Ontology-driven discourse analysis in
GenIE. In Proceedings of the 8th International Con-
ference on Applications of Natural Language to Infor-
mation Systems.
F. Ciravegna. 2001. Adaptive information extraction
from text by rule induction and generalization. In Pro-
ceedings of tht 17th International Joint Conference on
Artificial Intelligence (IJCAI 2001).
S. Davies, M. Poesio, F. Bruneseaux, and L. Ro-
mary, 1998. Annotating Coreference in Di-
alogues: Proposal for a scheme for MATE.
http://www.cogsci.ed.ac.uk/˜poesio/anno manual.html.
S. Decker, M. Erdmann, D. Fensel, and R. Studer. 1999.
Ontobroker: Ontology Based Access to Distributed
and Semi-Structured Information. In R. Meersman
et al., editors, Database Semantics: Semantic Issues
in Multimedia Systems, pages 351–369. Kluwer.
S. DeRose, E. Maler, and R. Daniel. 2001. XML Pointer
Language (XPointer). Technical report, W3C. Work-
ing Draft 16 August 2002.
S. Fligelstone. 1992. Developing a scheme for annotat-
ing text to show anaphoric relations. In G. Leitner, ed-
itor, New Directions in Corpus Linguistics, pages 153–
170. Mouton de Gruyter.
T.R. Gruber. 1993. Toward principles for the design
of ontologies used for knowledge sharing. In Formal
Analysis in Conceptual Analysis and Knowledge Rep-
resentation. Kluwer.
U. Hahn and K. Schnattinger. 1998. Ontology engineer-
ing via text understanding. In Proceedings of the 15th
World Computer Congress ’The Global Information
Society on the Way to the Next Millenium’ (IFIP’98).
S. Handschuh, S. Staab, and A. M¨adche. 2001. CREAM
— Creating relational metadata with a component-
based, ontology-driven annotation framework. In Pro-
ceedings of K-Cap 2001, pages 76–83. ACM Press.
L. Hirschman and N. Chinchor. 1997. MUC-7 corefer-
ence task definition. In Proceedings of the 7th Mes-
sage Understanding Conference (MUC-7).
Nancy Ide. 2002. Linguistic annotation framework.
Technical Report ISO TC37/SC4/WG1 N11, ISO
TC37 SC4 Language Resource Management, 08.
M. Kifer, G. Lausen, and J. Wu. 1995. Logical foun-
dations of object-oriented and frame-based languages.
Journal of the ACM, 42:741–843.
E. Krahmer and P. Piwek. 2000. Varieties of anaphora:
Introduction. In ESSLLI 2000 Reader.
A. Lascarides and N. Asher. 1991. Discourse relations
and defeasible knowledge. In Meeting of the Associa-
tion for Computational Linguistics, pages 55–62.
O. Lassila and R. Swick. 1999. Resource descrip-
tion framework (RDF) model and syntax specifica-
tion. Technical report, W3C. W3C Recommendation.
http://www.w3.org/TR/REC-rdf-syntax.
A. M¨adche, V. Pekar, and S. Staab. 2002. Ontology
learning part one - on discovering taxonomic relations
from the web. In Web Intelligence. Springer.
M. Marcus, B. Santorini, and M.A. Marcinkiewicz.
1993. Building a large annotated corpus of en-
glish: The penn treebank. Computational Linguistics,
19(2):313–330.
C. M¨uller and M. Strube. 2001. Annotating anaphoric
and bridging relations with MMAX. In Proceedings of
the 2nd SIGdial Workshop on Discourse and Dialogue,
pages 90–95.
R. Passoneau, 1996. DRAMA: Instructions for Applying
Reference Annotation for Multiple Applications.
M. Poesio and U. Reyle. 2001. Underspecification in
anaphoric reference. In Proceedings of the IWCS-4.
M. Poesio and R. Vieira. 1998. A corpus-based investi-
gation of definite description use. Computational Lin-
guistics, 24(2):183–216.
Philip Resnik. 1997. Selectional preference and sense
disambiguation. In Proceedings of the ACL SIGLEX
Workshop on Tagging Text with Lexical Semantics:
Why, What, and How?
S. Staab and A. M¨adche. 2000. Ontology engineering
beyond the modeling of concepts and relations. In
R. V. Benjamins, A. Gomez-Perez, N. Guarino, and
M. Uschold, editors, Proceedings of the 14th Confer-
ence on Artificial Intelligence, Workshop on Applica-
tions and Problem-Solving Methods.
Kees van Deemter and Rodger Kibble. 2000. On core-
ferring: Coreference in MUC and related annotation
schemes. Computational Linguistics, 26(4).
