An Approach for Resource Sharing in Multilingual NLP

Original · EN

In this paper we describe an approach for the analysis of documents in German and English with a shared pool of resources. For the analysis of German documents we use a document suite, which supports the user in tasks like information retrieval and information extraction. The core of the document suite is based on our tool XDOC. Now we want to exploit these methods for the analysis of English documents as well. For this aim we need a multilingual presentation format of the resources. These resources must be transformed into an unified format, in which we can set additional information about linguistic characteristics of the language depending on the analyzed documents. In this paper we describe our approach for such an exchange model for multilingual resources based on XML.