Cross-Lingual Information Retrieval and Semantic Interoperability for Cultural Heritage Repositories

Proceedings of Recent Advances in Natural Language Processing, pages 483–490, Hissar, Bulgaria, 7-13 September 2013. Cross-Lingual Information Retrieval and Semantic Interoperability for Cultural Heritage Repositories Monti Johanna University of Sassari jmonti@uniss.it Mario Monteleone University of Salerno mmonteleone@unisa.it Maria Pia di Buono University of Salerno dibuono@unisa.it Federica Marano University of Salerno fmarano@unisa.it Abstract This paper describes a computational linguis- tics-based approach for providing interopera- bility between multi-lingual systems in order to overcome crucial issues like cross-language and cross-collection retrieval. Our proposal is a system which improves capabilities of lan- guage-technology-based information extrac- tion. In the last few years various theories have been developed and applied for making multi- cultural and multilingual resources easy to ac- cess. Important initiatives, like the develop- ment of the European Library and Europeana, aim to increase the availability of digital con- tent from various types of providers and insti- tutions. Therefore the accessibility to these re- sources requires the development of environ- ments enabling to manage multilingual com- plexity. In this respect, we present a methodo- logical framework which allows mapping both the data and the metadata among the language- specific ontologies. The feasibility of cross- language information extraction and semantic search will be tested by implementing an early prototype system. 1 Introduction The growing need by users to access information on the web in languages different from their own is fostering the research in the field of Cross- language Information Retrieval (CLIR) applica- tions. Typically in state-of-the-art CLIR applica- tions, information is searched by means of a que- ry expressed in the user’s mother tongue. This query is automatically translated in the desired foreign language and the results are translated back in the user’s mother tongue. This process is based on two different transla- tion stages: query translation and document translation. The query translation concerns the translation in the desired foreign language of the query expressed in the user’s mother tongue, whereas the document translation is the back translation in the user’s language of the relevant documents found by means of the translated que- ry. CLIR success obviously depends on the quali- ty of translation and therefore inaccurate transla- tions may cause serious problems in retrieving the relevant information in a foreign language. A very frequent source of mistranslations in specific domain texts is represented by multi- word units (MWU). MWUs designate a wide range of lexical constructions, composed of two or more words with an opaque meaning, i.e. the meaning of a unit is not always the result of the sum of the meanings of the single words that are part of the unit. MWUs are not always easy to identify since co-occurrence among the lexemes forming the units may vary a great deal. A par- ticular type of MWUs are term compounds, i.e. various types of compounds, but mainly noun compounds, which belong to a language for spe- cial purposes (LSP). In all languages there is a close relationship between terminology and mul- ti-words and, in particular, word compounds. In fact, word compounds …

Content truncated to 500 words