=Paper=
{{Paper
|id=Vol-431/paper-21
|storemode=property
|title=Literature-based alignment of ontologies
|pdfUrl=https://ceur-ws.org/Vol-431/om2008_poster2.pdf
|volume=Vol-431
|dblpUrl=https://dblp.org/rec/conf/semweb/LambrixTX08
}}
==Literature-based alignment of ontologies==
https://ceur-ws.org/Vol-431/om2008_poster2.pdf
Literature-based alignment of ontologies
Patrick Lambrix and He Tan and Wei Xu
Department of Computer and Information Science
Linköpings universitet, Sweden
Abstract. In this paper we propose and evaluate new strategies for
aligning ontologies based on text categorization of literature using sup-
port vector machines-based text classifiers, and compare them with ex-
isting literature-based strategies. We also compare and combine these
strategies with linguistic strategies.
1 Introduction
In recent years many ontologies have been developed and many of these ontolo-
gies contain overlapping information. A number of ontology alignment systems
that support the user to find inter-ontology relationships exist (see overviews in
e.g., [2, 5] and http://www.ontologymatching.org/). Recently, there is a grow-
ing interest in instance-based methods for ontology alignment. In this paper we
slightly generalize the method for instance-based ontology alignment using lit-
erature that was proposed in [7]. Further, we propose a new instantiation of the
method based on text categorization using support vector machines (SVMs). We
evaluate these algorithms in terms of the quality of the alignment results for the
five test cases used in [7]. We compare two SVM-based algorithms with each other
and with the Naive Bayes text classification approach of [7]. Finally, we compare
the algorithms with a good text-based approach and discuss the advantages and
disadvantages of combining the approaches. For related work, more results and
more details we refer to the longer version of this paper that is available from
the SAMBO website (http://www.ida.liu.se/∼iislab/projects/SAMBO/).
2 Background
Many ontology alignment systems are based on the computation of similarity
values between terms in different ontologies and can be described as instantia-
tions of the general framework defined in [5]. An alignment algorithm receives as
input two source ontologies. The algorithm can include several matchers. These
matchers calculate similarities between the terms from the ontologies. Alignment
suggestions are then determined by combining and filtering the results generated
by one or more matchers. The suggestions are then presented to the user who
accepts or rejects them.
A method for creating a matcher that uses scientific literature was proposed
in [7]. It builds on the intuition that a similarity measure between concepts can
be computed based on relationships between the documents in which they are
used. It contains the following basic steps (slightly generalized). (1) Generate
corpora. For each ontology that we want to align we generate a corpus of doc-
uments. (2) Generating classifiers. For each ontology one or more document
classifiers are generated. The corpus of documents associated to an ontology is
used for generating its related classifiers. (3) Classification. Documents of one
ontology are classified by the document classifiers of the other ontology and vice
versa. (4) Calculate similarities. A similarity measure between concepts in
the different ontologies is computed based on the results of the classification.
In [7] an instantiation (NB) of this method was implemented and evaluated
using test cases involving biomedical ontologies. For step 1 a corpus was gener-
ated by querying PubMed (October 23, 2005) with each concept and retrieving
the 100 most recent abstracts (if there were so many) for each concept. In step
2 one Naive Bayes classifier per ontology was generated. The classifiers return
for a given document d the concept C in the ontology for which the posterior
probability P (C|d) results in the highest value. In step 3 the Naive Bayes clas-
sifier for one ontology was applied to every abstract in the abstract corpus of
the other ontology and vice versa. Finally, in step 4 a similarity value between
two concepts was computed using the numbers of abstracts associated with one
concept that are also related to the other concept as found by the classifiers.
In general, in step 2 a document may be assigned to several concepts and
thus we may regard the classification of documents to concepts as several binary
classification problems, one for each concept in an ontology. In the next section
we propose an instantiation of the method that does exactly this and is based
on SVMs. SVMs [8] is a machine learning method that constructs a separating
hyperplane in a feature space between two data sets (positive and negative ex-
amples) which maximizes the margin between the two sets. The setting can also
be generalized to learning from positive and unlabeled examples (e.g. [6]).
3 Alignment algorithms
The basic algorithm implements the steps as follows. (1) Generate corpora.
We used the same corpora as in [7]. (2) Generating the classifiers. For each
concept in each ontology an SVM text classifier was generated. We used the LPU
[6] system. LPU generates text classifiers based on positive and unlabeled exam-
ples. The abstracts retrieved when querying for a concept were used as positive
examples for that concept. Further, for a given concept we used one abstract
of each other concept in the same ontology as unlabeled examples. The SVM
text classifier for a concept returns for a given document whether the document
is related to the concept. It returns a value that is positive if the document is
classified to the concept and negative otherwise. (3) Classification. The SVM
text classifier for each concept in one ontology is applied to every abstract in the
abstract corpus of the other ontology and vice versa. The classification was done
by using the text classifiers generated by LPU within the SVMlight system [4].
Observe that a document can be classified to zero, one or more than one concept
in an ontology. (4) Calculate similarities. We define the similarity between a
concept C1 from the first ontology and a concept C2 from the second ontology as:
nSV MC−C2 (C1 , C2 ) + nSV MC−C1 (C2 , C1 )
nD (C1 ) + nD (C2 )
where nD (C) is the number of abstracts originally associated with C, and
nSV MC−Cq (Cp , Cq ) is the number of abstracts associated with Cp that are also
related to Cq as found by classifier SV M C − Cq related to concept Cq .
The pairs of concepts with a similarity measure greater or equal than a
predefined threshold are then presented to the user as candidate alignments.
In NB a document was classified to exactly one concept. We wanted to eval-
uate whether this has a real influence in the similarity computation. Therefore,
we also developed an alternative to the basic SVM algorithm where in step 3 a
document can be classified to only one concept. We assign a document only to
the concept for which its SVM classifier generated the highest positive value for
that document. In the case more than one classifier produces the highest positive
value, then one of the associated concepts is chosen.
4 Evaluation
We evaluate the proposed algorithms with respect to the quality of the sugges-
tions they generate. We also compare them to NB as well as to the best text-
based matcher (TermWN) implemented in SAMBO [5]. Further, we investigate
the combination of the proposed algorithms and TermWN.
We used the following set-up. We use the same five test cases as in [7]. For
the first two cases we use a part of a Gene Ontology (GO) ontology together
with a part of Signal Ontology (SigO). The first case, B (behavior), contains 57
terms from GO and 10 terms from SigO. The second case, ID (immune defense),
contains 73 terms from GO and 17 terms from SigO. The other cases are taken
from the anatomy category of Medical Subject Headings (MeSH) and the Adult
Mouse Anatomy (MA): nose (containing 15 terms from MeSH and 18 terms from
MA), ear (containing 39 terms from MeSH and 77 terms from MA), and eye
(containing 45 terms from MeSH and 112 terms from MA). Golden standards
for these cases were developed by domain experts. Further, we use the same
corpus as in [7]. We use SVM-based matchers based on sets of maximum
100 documents per concept. These matchers are denoted as SVM-P and SVM-S
where P and S stand for Plural (a document can be classified to several concepts)
and Single (a document can be classified to only one concept), respectively.
The results are given in table 1. The first column represents the cases and the
number of expected alignments for each case based on the golden standards. The
expected alignments are a minimal set of suggestions that matchers are expected
to generate for a perfect recall. The second column represents threshold values.
The cells in the other columns contain quadruplets a/b/c/d which represent the
number of a) suggestions, b) correct suggestions, c) wrong suggestions and d)
inferred suggestions, for a given case, matcher and threshold.
Comparison of single and plural assignment. The recall for the plural
assignment is much higher than the recall for the single assignment. This comes,
however, at a cost. The precision for the single assignment algorithm is much
higher than for the plural assignment algorithm. We see a real trade-off here:
find many expected alignments, but also get many wrong suggestions, or, find
few expected alignments, but receive almost no wrong suggestions.
Comparison of NB and SVM-S. These two single assignment algorithms
give relatively few suggestions but have high precision. However, NB gives al-
ways more suggestions than SVM for the same threshold. NB also always gives
suggestions, except for case ID and threshold 0.8, while SVM-S often does not
give suggestions. It is clear that SVM-S does not perform well with high thresh-
olds. In general, NB has slightly better recall than SVM-S, while SVM-S has
slightly higher precision than NB.
Th SVM-P SVM-S NB TermWN TermWN+ TermWN+
SVM-S SVM-P
B 0.4 387/4/258/125 0/0/0/0 4/2/1/1 58/4/22/32 4/4/0/0 156/4/84/68
4 0.5 306/4/203/99 0/0/0/0 2/2/0/0 35/4/13/18 4/4/0/0 52/4/19/29
0.6 225/4/148/73 0/0/0/0 2/2/0/0 13/4/4/5 0/0/0/0 21/4/7/10
0.7 130/3/79/48 0/0/0/0 2/2/0/0 6/4/0/2 0/0/0/0 7/4/1/2
0.8 36/0/22/14 0/0/0/0 1/1/0/0 4/4/0/0 0/0/0/0 4/4/0/0
ID 0.4 672/8/592/72 2/2/0/0 9/6/3/0 96/7/66/23 8/6/2/0 302/8/262/32
8 0.5 490/8/433/28 0/0/0/0 5/5/0/0 49/7/25/17 6/6/0/0 155/7/127/21
0.6 336/8/300/28 0/0/0/0 2/2/0/0 16/5/5/6 2/2/0/0 71/7/48/16
0.7 222/6/191/25 0/0/0/0 1/1/0/0 7/5/2/0 1/1/0/0 19/7/7/5
0.8 108/5/93/10 0/0/0/0 0/0/0/0 6/4/0/2 0/0/0/0 7/5/2/0
nose 0.4 155/7/124/24 5/5/0/0 6/5/1/0 48/7/37/4 9/7/2/0 80/7/66/7
7 0.5 120/7/91/22 4/4/0/0 6/5/1/0 28/7/18/3 7/7/0/0 58/7/47/4
0.6 85/7/60/18 2/2/0/0 5/5/0/0 8/6/2/0 6/6/0/0 31/7/47/4
0.7 58/6/45/7 0/0/0/0 5/5/0/0 6/6/0/0 4/4/0/0 11/7/4/0
0.8 34/6/27/1 0/0/0/0 3/3/0/0 6/6/0/0 1/1/0/0 6/6/0/0
ear 0.4 1224/24/1056/144 14/12/2/0 18/16/2/0 155/26/110/19 34/25/8/1 585/27/481/77
27 0.5 957/23/822/112 11/10/1/0 15/14/1/0 99/26/65/8 27/23/4/0 203/26/146/31
0.6 696/22/590/84 1/1/0/0 12/11/1/0 47/26/19/2 17/17/0/0 96/24/64/8
0.7 478/22/392/64 0/0/0/0 11/10/1/0 34/26/8/0 12/12/0/0 55/23/28/4
0.8 278/21/223/34 0/0/0/0 3/3/0/0 28/25/3/0 1/1/0/0 29/21/6/2
eye 0.4 2055/25/1926/104 7/7/0/0 25/18/7/0 135/26/100/9 28/23/5/0 643/25/568/50
27 0.5 1481/25/1366/90 4/4/0/0 18/17/1/0 74/23/44/7 21/20/1/0 272/25/221/26
0.6 957/25/860/72 0/0/0/0 14/14/0/0 33/22/10/1 16/16/0/0 138/24/101/13
0.7 612/24/539/49 0/0/0/0 10/10/0/0 24/21/3/0 7/7/0/0 54/21/27/6
0.8 344/23/290/31 0/0/0/0 3/3/0/0 22/20/2/0 0/0/0/0 25/21/4/0
Table 1. Results.
Comparison with and combination with other matchers. The table also
shows the quality of the suggestions of TermWN (from [5]), and the combinations
(sum, equal weight) of TermWN with SVM-P and SVM-S. TermWN has higher
recall than SVM-S and NB. It also has better recall than SVM-P for the case ear,
but for the other cases the recall is similar. TermWN has better precision than
SVM-P, but worse than SVM-S and NB. Almost all expected alignments were
found by at least one SVM or NB matcher and threshold at least 0.4. TermWN
with threshold 0.4 missed 1 expected alignment for ID, 1 for ear and 1 for eye.
The combination of TermWN and SVM-S gave perfect results for B and
thresholds 0.4 and 0.5. Otherwise, when it gave suggestions, the precision was
high. For thresholds 0.4 and 0.5, SVM-S worked as a filter on TermWN by
removing many wrong suggestions at the cost of no or few correct suggestions.
For higher thresholds too many correct suggestions were removed. For most cases
and thresholds the combination of TermWN and SVM-P gave better recall than
TermWN and SVM-P. The precision of the combination was higher than the
precision for SVM-P, but lower than the precision for TermWN. As shown in
the longer version of the paper, the precision for the combination could become
better than the precision for TermWN by using the double threshold filtering
technique of [1] while keeping the recall at the same level for most cases.
5 Conclusion
We have proposed SVM-based algorithms for aligning ontologies using literature.
We have shown that there is a trade-off between the single and plural assignment
methods regarding precision and recall. Further, SVM-S and NB obtained similar
results. The combinations of TermWN with SVM-S and with SVM-P lead to a
large gain in precision compared to TermWN and SVM-P, with still a high recall.
References
1. Chen B, Tan H and Lambrix P. 2006. Structure-based filtering for ontology align-
ment. Proceedings of the IEEE WETICE Workshop on Semantic Technologies in
Collaborative Applications, pp 364-369.
2. Euzenat J and Shvaiko P. 2007. Ontology Matching. Springer.
3. Joachims T. 1998. Text Categorization with Support Vector Machines: Learning
with Many Relevant Features. Proceedings of the European Conference on Machine
Learning, LNCS 1398, 137-142.
4. Joachims T. 1999. Making Large-Scale SVM Learning Practical. Advances in Kernel
Methods - Support Vector Learning, B Schölkopf and C Burges and A Smola (eds),
MIT-Press. http://svmlight.joachims.org/
5. Lambrix P and Tan H. 2006. SAMBO - A System for Aligning and Merging Biomed-
ical Ontologies. Journal of Web Semantics, 4(3):196-206.
6. Liu B, Dai Y, Li X, Lee WS and Yu Ph. 2003. Building Text Classifiers Using Positive
and Unlabeled Examples. Proceedings of the Third IEEE International Conference
on Data Mining, 179-188. http://www.cs.uic.edu/∼liub/LPU/LPU-download.html
7. Tan H, Jakoniene V, Lambrix P, Aberg J and Shahmehri N. 2006. Alignment of
Biomedical Ontologies using Life Science Literature. Proceedings of the International
Workshop on Knowledge Discovery in Life Science Literature, LNBI 3886, 1-17.
8. Vapnik V. 1995. The Nature of Statistical Learning Theory. Springer.