Abstract
One way to achieve semantic interoperability when data lexicalized in different languages is by means of cross-lingual linking. Translation resources are used as an intermediate step to reduce the language barriers. The key challenge is to select the correct mapping among candidate matches. We define an experiment to study upper bounds for the correctness of a cross-lingual ontology matching system. We highlight different lexical characteristics that can support the selection step. We believe that our findings can be useful in the design of cross-lingual mapping algorithms.
Author supplied keywords
Cite
CITATION STYLE
Helou, M. A., & Palmonari, M. (2015). Upper bound for cross-lingual concept mapping with external translation resources. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9103, pp. 424–431). Springer Verlag. https://doi.org/10.1007/978-3-319-19581-0_41
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
