=Paper=
{{Paper
|id=Vol-360/paper-9
|storemode=property
|title=AceWiki: Collaborative Ontology Management in Controlled Natural Language
|pdfUrl=https://ceur-ws.org/Vol-360/paper-5.pdf
|volume=Vol-360
|dblpUrl=https://dblp.org/rec/conf/semwiki/Kuhn08
}}
==AceWiki: Collaborative Ontology Management in Controlled Natural Language==
https://ceur-ws.org/Vol-360/paper-5.pdf
AceWiki: Collaborative Ontology Management
in Controlled Natural Language
Tobias Kuhn
Department of Informatics, University of Zurich, Switzerland
tkuhn@ifi.uzh.ch
http://www.ifi.uzh.ch/cl/tkuhn
Abstract. AceWiki is a prototype that shows how a semantic wiki using
controlled natural language — Attempto Controlled English (ACE) in
our case — can make ontology management easy for everybody. Sen-
tences in ACE can automatically be translated into first-order logic,
OWL, or SWRL. AceWiki integrates the OWL reasoner Pellet and en-
sures that the ontology is always consistent. Previous results have shown
that people with no background in logic are able to add formal knowledge
to AceWiki without being instructed or trained in advance.
1 Introduction
Since ontologies are often defined within communities, semantic wikis could be
used for their collaborative creation and management. Unfortunately, most of
the existing semantic wikis do not support expressive ontology languages in a
general way. They do not allow the users to add complex axioms like “every
landlocked country borders no sea”. Furthermore, the existing semantic wikis
are often hard to understand for people who are not familiar with the technical
terms of logic and ontologies.
AceWiki1 tries to solve both problems by using controlled natural language.
Ordinary people who have no background in logic should be able to understand,
modify, and extend the formal content of a wiki.
Many existing semantic wikis are classical wikis enriched with semantic an-
notations. The goal is not to manage stand-alone ontologies, but rather to give
some kind of formal backbone to the wiki articles. We follow a different approach
— similar e.g. to the myOntology project [7] — by providing a wiki that is dedi-
cated to building and maintaining ontologies. In contrast to myOntology, we do
not restrict ourselves to lightweight (i.e. relatively inexpressive) ontologies. The
use of controlled natural language allows us to express also complex axioms in
a natural way. Figure 1 shows a screenshot of the AceWiki interface.
In our usage scenario, a community of domain experts uses AceWiki to cre-
ate and maintain a formal knowledge base in a collaborative manner. There are
two exemplary wiki instances — one about geography and the other about pro-
tein interactions — that demonstrate how AceWiki could be used to represent
knowledge of such communities.
1
See [6] and http://attempto.ifi.uzh.ch/acewiki
Fig. 1. The web interface of the AceWiki prototype
AceWiki has been introduced in [6]. Since then, several new features have
been added, for example the integration of a reasoner and the support for number
restrictions (“at most 3”, “exactly 5”, etc.).
2 Attempto Controlled English
Attempto Controlled English (ACE)2 is the controlled natural language that is
used for AceWiki. ACE appears completely natural since it is a subset of English.
Restrictions of the syntax and the definition of a small set of interpretation rules
make it a formal language that is automatically translatable into first-order logic.
ACE supports a wide range of natural language constructs: singular and plural
noun phrases, active and passive voice, relative phrases, anaphoric references,
existential and universal quantifiers, negation, modality, and more. In the past,
ACE has successfully been applied for different tasks in different research areas,
for example as a query language for ontologies [1], as a knowledge representation
language for the biomedical domain [4], and as a rule language for a multi-
semantics rule engine [5].
Furthermore, ACE has been used as a natural language front-end to OWL
with a bidirectional mapping of ACE to OWL [3]. This mapping covers all
of OWL 2 except data properties and some very complex class descriptions.
AceWiki relies on this work for translating ACE sentences into OWL, which
allows us then to do reasoning with existing OWL reasoners.
2
See [2] and http://attempto.ifi.uzh.ch
Fig. 2. The predictive editor of AceWiki
3 Design and Evaluation
The goal of AceWiki is to show that semantic wikis can be more natural and at
the same time more expressive than existing semantic wikis.
Naturalness is achieved by representing the formal statements in ACE. Since
ACE is a subset of natural English, every English speaker can immediately read
and understand the content of the wiki. In order to enable easy creation of
ACE sentences, AceWiki provides a predictive editor that shows step-by-step
the words that are syntactically possible at a given position in the sentence.
Figure 2 shows a screenshot of the predictive editor of AceWiki. Furthermore,
the AceWiki interface does not use technical terms like “ontological element”,
“property”, or “subclass” but uses instead terms like “word”, “transitive verb”,
or “hierarchy” which should be much more familiar to people with no background
in logic.
AceWiki makes use of the high expressivity of ACE that goes beyond OWL
and SWRL. We do not like the idea of cutting down the expressivity just for
the sake of reasoning performance. Even if some statements become so complex
that it is almost impossible to do reasoning with them, it is better to have them
formalized than just left out. We do not lose anything, since we are free to ignore
those complex statements for certain reasoning tasks.
In our previous work [6], we conducted a user experiment that proved that
ordinary people with no background in logic are able to deal with AceWiki. The
participants — without being instructed how to interact with the interface —
were asked to add knowledge to AceWiki. About 80% of the created sentences
were correct and sensible. This is remarkable since most of the sentences were
quite complex: more than 60% of them contained an implication or a negation or
both. Using the predictive editor which the participants had never seen before,
they needed on average only five minutes to create their first correct sentence.
4 Reasoning in AceWiki
We have started to integrate the OWL reasoner Pellet3 into AceWiki. Since ACE
sentences can be beyond the expressivity of OWL, the reasoner cannot consider
all sentences. In order to make this clear to the users, each sentence is tagged as
blue (inside of OWL) or red (outside of OWL):
In this way, it is easy to explain to the users that only the blue statements are
considered when the reasoner is used. We plan to provide an interface that allows
skilled users to export the formal content of the wiki and to use it within an
external reasoner or rule-engine. Thus, even though the red statements cannot
be interpreted by the built-in reasoner they can still be useful.
Consistency checking plays a crucial role because any other reasoning task re-
quires a consistent ontology in order to return useful results. Most other semantic
wikis do not have this problem since their languages are simply not expressive
enough to ever run into inconsistency.
In order to ensure that the ontology is always consistent, AceWiki checks
every new sentence — immediately after its creation — whether it is consis-
tent with the current ontology. Otherwise, the sentence is not included in the
ontology:
After the user created the last sentence of this example, AceWiki detected that
it contradicts the current ontology. The sentence is included in the wiki article
but the red font indicates that it is not included in the ontology. The user can
remove this sentence again, or keep it and try to reassert it later when the rest
of the ontology has changed.
For this approach, it is very important to perform incremental reasoning
which Pellet supports only partially at the moment. For that reason, AceWiki
does not scale very well. We expect that future reasoners will be able to run
much faster in such incremental scenarios.
3
http://pellet.owldl.com/
Not only asserted but also inferred knowledge can be represented in ACE. At
the moment, AceWiki can show inferred class hierarchies and class memberships.
Furthermore, we are working on a query feature for AceWiki. Questions will be
formulated in ACE and evaluated by the reasoner:
Thus, ACE can be used not only as an ontology- and rule-language, but also as
a query-language.
5 Conclusions
The AceWiki prototype shows how ontologies can be managed in a natural
way within a wiki. It demonstrates how semantic wikis using controlled natural
language can be expressive and easy to use at the same time. Our previous eval-
uation showed that AceWiki is indeed easy to learn. We explained how AceWiki
ensures — in a very simple way — the consistency of the ontology which is the
basis for other integrated reasoning services.
References
1. Abraham Bernstein, Esther Kaufmann, Norbert E. Fuchs, June von Bonin. Talking
to the Semantic Web — A Controlled English Query Interface for Ontologies. Proc.
14th Workshop on Information Technology and Systems, 2004
2. Norbert E. Fuchs, Kaarel Kaljurand, Gerold Schneider. Attempto Controlled
English Meets the Challenges of Knowledge Representation, Reasoning, Inter-
operability and User Interfaces. Proc. 19th International FLAIRS Conference
(FLAIRS’2006), 2006
3. Kaarel Kaljurand. Attempto Controlled English as a Semantic Web Language. PhD
thesis, Faculty of Mathematics and Computer Science, University of Tartu, 2007
4. Tobias Kuhn, Loı̈c Royer, Norbert E. Fuchs, Michael Schroeder. Improving Text
Mining with Controlled Natural Language: A Case Study for Protein Interactions.
Proc. Third International Workshop on Data Integration in the Life Sciences 2006
(DILS’06), Springer, 2006
5. Tobias Kuhn. AceRules: Executing Rules in Controlled Natural Language. Proc.
First International Conference on Web Reasoning and Rule Systems (RR2007),
Springer, 2007
6. Tobias Kuhn. AceWiki: A Natural and Expressive Semantic Wiki. Proc. of Seman-
tic Web User Interaction at CHI 2008: Exploring HCI Challenges, CEUR Work-
shop Proceedings, 2008
7. Katharina Siorpaes, Martin Hepp. myOntology: The Marriage of Ontology Engi-
neering and Collective Intelligence. Proc. Bridging the Gap between Semantic Web
and Web 2.0, 2007