The paper presents a novel technique for aligning cross-lingual ontologies that does not rely on machine translation, but uses the large multilingual semantic network BabelNet as a source of background knowledge. In addition, our approach applies a novel orchestration of the components of the matching work ow. We demonstrate that our method outperforms considerably the best techniques in the state-of-the-art.
1.1
We present LYAM++ (Yet Another Matcher - Light), a fully automatic crosslingual ontology matching system that does not rely on machine translation. Instead, we make use of the openly available general-purpose multilingual semantic network BabelNet1 in order to recreate the missing semantic context in
the matching process. Another original feature of our approach is the choice of
orchestration of the matching work ow. Our experiments on the MultiFarm2
benchmark data show that (
The work ow of LYAM++ is given in Fig. 1. We take as an input a source ontology S, given in a natural language lS and a target ontology T , given in a language lT . The overall processes consists of four main components: a terminological multilingual matcher, a mapping selection module and, nally, a structural matcher. One of the original contributions of this work is the choice of orchestration of these components. Indeed, the places of the mapping selection module and the structural matcher are reversed in the existing OM tools [4]. However, we wanted to ensure that we feed only good quality mappings to the structural matcher, therefore we decided to lter the discovered correspondences right after producing the initial alignment. This decision is supported experimentally in the following section.
The multilingual terminological matching module, the second contribution described in this paper, acts on the one hand as a preprocessing component and, on the other hand { as a light-weight terminological matcher between crosslingual labels. We start by splitting the elements of each ontology in three groups: labels of classes, labels of object properties and labels of data object properties (in colors blue, black and red in the gure), since these groups of elements are to be aligned separately. A standard preprocessing procedure is applied on these sets of labels, comprising character normalization, stop-words ltering, tokenization and lemmatization. The tokens of the elements of T are then aligned to BabelNet. At rst, every token of a given label s in S is enriched by related
terms and synonyms from BabelNet and all of these terms are represented in the language lT , which makes these terms comparable to the tokens of the labels in T . A simple similarity evaluation by the help of the Jaccard coe cient selects the term in each set of related terms corresponding to a given token from s that has the highest score with respect to every token in each label of T . This helps to restitute the label s in the language lT . Finally, the labels in each group of S and T , seen as sets of tokens, are compared by using the Soft TFIDF similarity measure [5], which produces an intermediate terminological alignment.
The three remaining components are standard OM modules [4], although ordered in a new manner. The Mapping selection is a module that transforms the initial 1 to many mapping to a 1:1 alignment based on the principle of iteratively retaining the pairs of concepts with maximal value of similarity. Finally, the structural matcher component lters the trustworthy pairs of aligned concepts by looking at the similarity values produced for their parents and their children in the ontology hierarchies. 1.3
The system is not yet available online. The reason for that is that it depends heavily on the use of BabelNet, which is a protected source. We are working on implementing a sharable version of LYAM++ making use of di erent open access background knowledge sources. 1.4
The alignments produced by LYAM++ for this year's Multifarm track can be found under the following link: http://www.lirmm.fr/benellefi/Lyam++.rar 2
We have evaluated our approach on data coming from the ontology alignment evaluation initiative (OAEI)3 and particularly Multifarm|a benchmark designed for evaluating cross-lingual ontology matching systems. Multifarm data consist of a set of 7 ontologies originally coming from the Conference benchmark of OAEI, translated into 8 languages. Two evaluation tasks are de ned: task 1 consists in matching two di erent ontologies given in di erent languages, while task 2 aims to align di erent language versions of one single ontology.
We have performed experiments on both tasks by using the pairs of languages given in the summary of our results in Table 1.
In another experiment, we have evaluated the results obtained by using our novel orchestration of matching components, as compared to the standard orchestration. The gures in Table 2 show that the work ow proposed in this paper acts in favor of achieving better results as compared to the standard method.
Lang. pair LYAM++ 0.54 0.58 0.62 0.60 0.60 0.60 0.63 0.67 0.53 Average F-measures over all threshold values per language pair for task 1.
FR-RU FR-PT FR-NL ES-FR ES-RU ES-PT ES-NL EN-PT EN-RU EN-FR
Discussions on the way to improve the proposed system Currently, we are working on enhancing the system in order to make applicable to the general ontology matching problem and not only to cross-lingual ones. We have generated rst results on the Conference benchmark without any modi cation in the settings and our results are quite promising. For the majority of the datasets (ontology pairs) our system achieves a f-score almost as good as the f-score of AML, the best performing system on that track.
We consider that a key feature for the improvement of our system is the appropriate choice of background knowledge. In order to improve the results achieved on the Conference track, we plan to use monolingual general purpose background knowledge (for example, the english subgraphs of YAGO or DBPedia) instead of BabelNet.
We intend to use domain speci c background knowledge in order to solve alignment problems in speci c areas of knowledge. More precisely, we plan to participate on the Anatomy track by testing di erent kinds of domain speci c background knowledge, such as UMLS or other. 4
We presented an e cient approach for aligning cross-lingual ontologies by
using the multilingual lexical database BabelNet. Subjects of ongoing and future
work are (