Towards a new generation of Language Resources: global trends and international convergences

Summary form only given. Language resources (LR) and semantics, dynamic resources automatically acquired, interoperability among LRs and Language Technology (LT), and how to go for a new generation of LRs in the Semantic Web (SW) framework, pointing at the potentialities and the need of a cross-fertilisation between the two communities of Human Language Technology (HLT) and SW. Large scale LRs are unanimously recognised as the necessary infrastructure underlying LT. This paper explains about how computational lexicons and textual corpora should be considered as complementary views on the lexical space, in the perspective of modelling a new type of resource which is both a lexicon and a corpus together. A ´complete´ computational lexicon should incorporate and represent our ´knowledge of the world´. We claim that it is theoretically impossible to achieve completeness within any ´static´ lexicon. Moreover, choices on the syntagmatic axis are pervasive in language.