Duration of the project: 2003-2006. The project proposes to continue the work in the processing of the natural language from scientific-technical corpus that since 1994 the Group of research in Lexicon, Terminology and Specialized Speech (IULATERM) is carrying out and for which it has had help of the National Plan, of the Plan of the Autonomous Community and of the Union European. The goals that are intended with this project are divided up in theoretical-applied and applied-technological. More the suggested goals are precisely, in the theoretical ground orientated to the automatic extraction of information: a) the analysis of the different types of Units of Specialized Knowledge (USKs) pertinent in the structure of knowledge of the texts and b) the analysis of the units that express the relation that, in the texts, establish among the USKs. In order to elaborate technological applications, the suggested goals are, in the applied ground: a) enriching the dictionaries of processing with not flexive morphological information (internal morphological structure), syntactic (argumentative structure) and semantic (semantic characteristics of the arguments); b) developing chunker of the internal structure of the lexicon (morphological chunker); c) developing a syntactic chunker of second level (syntactic chunker); and d) to improve the representation system automatic of the structure of knowledge of a text (conceptual maps). Duration of the project: 2003-2006. Besides the members of the group IULATERM, other members of the IULA collaborate.