Drs. T. GaustadDrs. P. Berck, K. van der Sloot, Drs. H. Stehouwer, Prof. dr. A. van den Bosch
Natural language processing models and systems typically employ abstract linguistic representations (syntactic, semantic, or pragmatic) as intermediate working units. The question underlying the Implicit Linguistics project is whether we can do without them, since any invented intermediate structure is always implicitly encoded somehow in the words at the surface, and the way they are ordered. Classes of natural processing tasks in which this question can be investigated in extremo are processes in which form is mapped to form, i.e., in which neither the input nor the output contains abstract elements to begin with. The Implicit Linguistics project focuses on spelling correction (converting a corrupted text to a clean, intended form of the same text) and on machine translation and paraphrasing (converting a text in one language to a text in the same or another language carrying approximately the same meaning).