had the objective of going a step ahead in discourse, grammatical and semantic analyses of specialized texts and lexical and phraseological units, in order to achieve an automatic extraction system of cognitive structures underlying specialized texts. The research was organized around four axes: text analysis and elements for tipologization; Units of specialized knowledge and concept representation; concept relations; information user needs. At the end of the proyect, a publication wil be done based on the various tecnical and descriptive reports. Project's term: 2001-2003. Besides the member of the group IULATERM, other members of IULA have taken part in this project.