Cross-Lingual Ontology Mapping – An Investigation of the Impact of Machine Translation

Abstract

Ontologies are at the heart of knowledge management and make use of information that is not only written in English but also in many other natural languages. In order to enable knowledge discovery, sharing and reuse of these multilingual ontologies, it is necessary to support ontology mapping despite natural language barriers. This paper examines the soundness of a generic approach that involves machine translation tools and monolingual ontology matching techniques in cross-lingual ontology mapping scenarios. In particular, experimental results collected from case studies which engage mappings of independent ontologies that are labeled in English and Chinese are presented. Based on findings derived from these studies, limitations of this generic approach are discussed. It is shown with evidence that appropriate translations of conceptual labels in ontologies are of crucial importance when applying monolingual matching techniques in cross-lingual ontology mapping. Finally, to address the identified challenges, a semantic-oriented cross-lingual ontology mapping (SOCOM) framework is proposed and discussed.

A Machine Learning Approach to Multilingual and Cross-Lingual Ontology Matching

“…We believe that e.g. what Fu et al [3] refer to as "multilingual ontologies" can thus be described more accurately as a cross-lingual matching scenario involving two (or more) monolingual ontologies.…”

“…[3,6,10]). In the following, we will define these notions, based on the general definition of ontology matching as "the process of finding relationships or correspondences between entities of different ontologies" (cf.…”

“…Fu et al [3] have argued that the matching quality in cross-lingual scenarios strongly depends on the translation quality of label translations generated by machine translation (mt) tools. This certainly holds for the present study as well, since the choice of the mt system determines e.g.…”

“…whether the Italian term "Conto economico" is translated as "Income statement" 5 or as "I count economical" 6 . The conclusion of Fu et al [3] is that good translation quality is a prerequisite for achieving good quality cross-lingual ontology matches. While we do believe that this is true when comparing cross-lingual ontology matching to monolingual ontology matching with high-quality labels, it is worth investigating the possibility of translating the labels of both ontologies to a third language, as it may be the case that the quality difference between the labels can thus be reduced.…”

“…3 However, while these algorithms generally focus on using information in a single language that is shared by two ontologies, multilingual and cross-lingual aspects of ontology matching arewith the notable exception of [3,10] -still largely understudied. Assuming that more and more ontological resources will become available on the Semantic Web in more than one language, it is necessary to develop novel algorithms which are able to leverage multilingual information in case it is available, as well as capable of bridging the gap in case two ontological resources which do not share any languages need to be matched.…”