Terminology and natural language processing are at the basis of many computer applications dealing with domain knowledge, from text generation, information retrieval, machine-assisted translation, and ontology-based language annotation, to ontology localization to mention just a few. In all these areas, the final aim is to allow people to communicate with computers. This communication can be seen from a wide perspective in the new Semantic Web where ontologies have achieved its highest point as the knowledge representations that can be shared by users and machines. In this respect, one of the most challenging issues nowadays is the task of adapting domain ontologies to users of different languages and cultures while preserving the generality of a domain model. Another important issue is derived from the ever-increasing need of improving cross-lingual information extraction methods by means of new innovative approaches that exploit web-based resources (lexica, corpora, translation services, etc.).The linguistic model (Linguistic Information Repository, LIR) developed in our group to represent the linguistic knowledge necessary in ontology localization as well as the tool that implements this model and semi-automates the ontology localization process (LabelTranslator) are two important advances that will be further improved in the near future.
Apart from the research developed in other areas, such as natural language generation, ontological formalization of linguistic resources, standardization of linguistic annotation tag sets, and interoperation of linguistic annotation tools and other annotation linguistic resources by means of ontologies, our team is now involved in a new EU project, MONNET, Multilingual Ontologies for Networked Knowledge (FP7-248458). At this stage, our aim is to obtain a new model integrating different linguistic and multilingual aspects that up to now have been partially approached. We also aim at offering an automatic and adaptive localisation of ontologies by building on and enhancing available models for associating linguistic and multilingual data to ontologies.
Some of the innovative aspects will deal with scalability and adaptability to new application domains and languages through a web-based approach that will leverage the increased availability of linguistic resources as well as semantic and linked data.
This research area is led by Guadalupe Aguado de Cea, Associate Professor, and the team is formed by the following members:
Some of the most important publications in these areas are:
There are currently no job offers or studentships available in this research area. For offers in other areas of the group, please check in our job opportunities section.
Created under Creative Commons License - 2010 OEG.