Development of formalized functional-synonymic models of some nominative French and Russian structures

The article considers the problems of creation separate functional-semantic models during transformation of nominative structures for solving the text analysis tasks and presenting these structures in formalized descriptions of multilingual knowledge base Intertext. The same formal descriptions are used as a basis for accomplishing automated linguistic transformations in machine translation. For some non-prepositional, prepositional and prepositional-casal nominative structures the authors have developed the basic formalized semantic-synonymic Russian and French constructions. These models are being developed within the framework of the IPI RAN project on creation of multilingual knowledge base Intertext and multilingual linguistic processor. The analysys of structures and the module of multivariant formal-logical rules are being developed in view of functional similarity and syntactic polysemanticity. At the moment, the module of French-Russian (and reverse) translation is being developed for the current experimental version of the processor basing on English and Russian. At the first stage, the analysis of nominative structures based on the targeted sampling of parallel scientific texts in Russian and French is to be done. The task of meaning transfer is solved on the basis of the analysis of cognitive structures.