=Paper=
{{Paper
|id=Vol-2032/oaei17_paper9
|storemode=property
|title=ONTMAT: results for OAEI 2017
|pdfUrl=https://ceur-ws.org/Vol-2032/oaei17_paper9.pdf
|volume=Vol-2032
|authors=Saida Gherbi,Mohamed Tarek Khadir
|dblpUrl=https://dblp.org/rec/conf/semweb/GherbiK17
}}
==ONTMAT: results for OAEI 2017==
https://ceur-ws.org/Vol-2032/oaei17_paper9.pdf
ONTMAT: Results for OAEI 2017
Saida Gherbi1and MedTarek Khadir2
1
LabGed, ESTI, Annaba 23000, Algeria
Saida_gharbi23@yahoo.fr
2
LabGed, University Badji Mokhtar Annaba, 23000, Algeria
Khadir@labged.net
Abstract: This paper describes ONTMAT an ontology matching system, and
presents the results obtained for the Ontology Alignment Evaluation Initiative
(OAEI) 2017. ONTMAT is an ontology matching process, which compares the
instances of ontologies to align in order to deduce the relations between their
concepts. Then, based on hierarchical and binary relations between the concepts
inside the ontologies it performs entities matching.
Keywords: Ontology, Alignment, OWL.
1 Presentation of the system
ONTMAT (ONTology MATching) is an ontology alignment tool, aiming to align
OWL entities (classes, object properties i.e. binary relations), participating for the first
time in OAEI (Conference track).
1.1 State, purpose, general statement
ONTMAT uses a terminological methods based on WordNet dictionary [2],which is
exploited as background knowledge to provide a set of the relations between individ-
uals names of the ontologies source ( ) and target( ). Then, if the name does not
exist in WordNet the approach handles the n-gram measure instead of the dictionary.
Moreover, from this set of individual relations we will deduce the equivalence or
subsumption relation among their concepts. The equivalent concepts are recorded in
an alignment matrix (AM), and the concepts related by subsumptions relations are
registered within a temporary alignment matrix (TAM) [4].
Furthermore, the TAM elements and the concepts neighbors of AM are compared
by using the inference roles with the terminological techniques cited previously and
the retained alignment will be added to AM. The concepts neighbors are those related
by hierarchical or binary relations with AM concepts. Here, we first align the neigh-
boring concepts because they have more chance to be similar [1], after we will align
the other concepts by using the same technics. Next, inference technics are applied on
AM to align the binary relations.
2
1.2 Approach description
In our proposition we suppose that Wordnet is hierarchically organized as
W(S,≤,Ag,g), where S is a set of synsets {s1,s2,…,si}(i is a positive integer), and an
annotate function Ag associates the gloss g to each synset. Furthermore, the relations
≤ between concepts , may be presented in the following logical relations [4] as:
1) ;means that is a hyponym or meronym of ;
2) ; express that is a hypernym or holonym of ;
3) ; signified that and belong to the same synset are similar[1];
4) when and are the siblings in the part of hierarchy they are con-
nected by a relation of antonymy.
The entities aligned can be related by one of the hierarchical relation presented in
the set , , where ( : equivalence; subclass), fuzzy relation symbol-
ized by “&”, or binary relation. Further, the binary ontologies relations (O1, O2)are
also aligned by an element of the set HR. The algorithm will explain in the following
items:
1. In level 1 we compare the instances names (IO1,IO2) of ontologies( , )to deduce
the relations among their concepts. To do this, WordNet is exploited because we
cannot assume with certainty that two entities are dissimilar if they have different
names (synonyms), or they are equivalent if they have the same name(homonyms).
If the name does not existed in WordNet we will measure the similarity among
names by the n-gram measure. Then the equivalent instances will construct the in-
stances matrix IM. The concepts (C1,C2) of ( , ) that have the same sets of in-
stances in IM are considered as equivalent concepts as proven in [4], and
( , , ) can be added to AM. Although, if the instances set of C1 are included in
the instances set of C2then ( , , ) will be inserted in TAM.
2. The level 2 starts by applying terminological techniques on the concepts names.
Next the results obtained will be combined with inferences methods illustrated in
[4] to be inserted in AM: ( , , ) of TAM confirmed or modified, where
, , , .
3. The concepts neighbors sorted by hierarchical relations of the AM elements
( , , ), are ( , linked to ( , respectively by an element of the set
HR. The neighbors , joined by " " with , of AM in ( , ), will be
aligned by using the inferences techniques applied on the background knowledge
[3]. The background knowledge is the ontology source when the neighbors belong
to O1, and ontology target if we match the neighbors existing in O2. The other
neighbors will be matched using the terminological methods.
4. The fourth level exploits the description logic roles proven in [4] to match the con-
cepts ( , associated to ( , by binary relations ( , ) in ( , ), as fol-
lowing:
─ If ( , , ( , , then ( , , will be inserted to ABRSM (Align-
ment Binary Relation Source Matrix)
─ If ( , , ( , , then ( , , will be added to ABRTM (Alignment
Binary Relation Target Matrix)
3
Thus, binary relations can be aligned because we have: hR Iff dom( ) hR
dom( ) and ran( ) hR ran( ); where [4], for instance;
If ( , , , , , , ( , , then ( , , ) will
be added to ABR (Alignment Binary Relation Matrix).
5. Finally, the concepts not yet aligned, will be matched via the terminological meth-
ods.
1.3 Adaptations made for the evaluation
We have adapted the format of the alignment result to the reference alignments re-
stricted to name classes, using the “=” sign for equivalence relation with confidence
of 1. Although our system provides other relations as subsumption, and binary rela-
tions without measure, as well as the alignment of binary relation by the HR.
2 Results
In this version we wish to test the techniques used by ONTMAT, such as, the infer-
ences mechanisms applied onWordNet and the ontologies source and target, and the
deduction of the matching among entities based on instances. The most appropriate
track to do these tests is the conference track.
Conference track comprises 16 ontologies from the domain of conference organiza-
tion. Most ontologies of this track were equipped with OWL DL axioms; which is
useful to test our inferences approach. Table 1shows the evaluation result obtained by
running ONTMAT under the SEALS client with the command:
java -jar F:/temp/seals-omt-client.jar F:/temp/ONTMAT -t
This command tests two predefined ontologies from the Conference. From Table 1
we can write that ONTMAT perform well because these ontologies are the same
structure.
Table 1. Results for two predefined ontologies
Precision Recall F-Measure
1.0 0.455 0.625
The results obtained by the global test as illustrated in Table 2, are not well as the
results of the precedent table in term of precision and F-measure. Although, the global
recall is 0.434.
Table 2.Results for conference track
Test Case ID Precision Recall F-measure
cmt-conference 0.6 0.2 0.3
cmt-confof 0.4 0.25 0.308
cmt-edas 0.444 0.615 0.516
4
cmt-ekaw 0.217 0.455 0.294
cmt-iasted 0.143 1.0 0.25
cmt-sigkdd 0.176 0.5 0.26
conference-confof 0.052 0.467 0.094
conference-edas 0.052 0.412 0.092
conference-ekaw 0.059 0.32 0.1
conference-iasted 0.03 0.286 0.054
conference-sigkdd 0.059 0.533 0.106
confof-edas 0.04 0.421 0.073
confof-ekaw 0.04 0.4 0.073
confof-iasted 0.02 0.444 0.038
confof-sigkdd 0.02 0.571 0.039
edas-ekaw 0.016 0.217 0.03
edas-sigkdd 0.022 0.467 0.042
ekaw-iasted 0.015 0.6 0.029
ekaw-sigkdd 0.017 0.636 0.033
iasted-sigkdd 0.02 0.733 0.039
Global 0.034 0.434 0.063
2.1 Discussions on the way to improve the proposed system
To improve our application, we will also align the properties of ontologies
( , ). Then, adapt it to read all files type, and integrate the translator to test our
tool under other tracks as: Instance Matching, MultiFarm.
2.2 Comments on the OAEI test cases
The application seals-omt-client from seal, only test files where the alignment rela-
tion between concepts is itself the equivalence relation. However ONTOMAT, offers
other possibilities in terms alignment relations between entities such as; & : Fuzzy
and binary relations. We hope that OAEI takes into consideration those types of rela-
tions in the reference alignment file.
3 Conclusion and future work
We have briefly described the mechanisms exploited by our proposition ONTMAT,
and presented the results obtained under the conference track of OAEI 2017.
This is our first participation in OAEI, the results are not satisfying, and the system
presents some limitations. In the future, we will make great efforts to improve
ONTMAT results, and participate in more tracks.
5
References
1. Euzenat, J., Shvaiko. P. “Ontology Matching” , pages 37-39; 73-87, 92-93. Springer-
Verlag Berlin Heidelberg, 2007.
2. Fellbaum, C. :WordNet: An Electronic Lexical Database, MIT Press, Cambridge, MA
(1998).
3. GherbiS., BelleiliH.andKhadirM., 2013. BRMAP : Un outil d’Alignement des ontologies.
7ème édition de la conférence maghrébine sur les avancés des systèmes décisionnels.
4. GherbiS., 535 M. T. Khadir, Inferred ontology concepts alignment using and an external
dictionary, Procedia Computer Science 83 (2016) 648- 652, the 7th International Confer-
ence on Ambient Systems, Networks and Technologies (ANT 2016) / The 6th Internation-
al Conference on Sustainable Energy Information Technology (SEIT-2016) / Aliated
Workshops.doi:http://dx.doi.org/10.1016/j.procs.2016.04.145.URL
http://www.sciencedirect.com/science/article/pii/ S1877050916301752.
5. Giunchiglia, F., Shvaiko, P., Yatskevich, M., 2004. SMatch an algorithm and an imple-
mentation of semantic matching. In Proc. 1st EuropeanSemanticWeb Symposium (ESWS),
volume 3053 of Lecture notes in computer science, pages 61–75, Hersounisous (GR), 10-
12 May 2004.