From Czech morphology through partial parsing to disambiguation

Abstract

This paper deals with a complex system of processing raw Czech texts. Several modules were implemented which perform different levels of processing. These modules can easily be incorporated into many other linguistic applications and some of them are already exploited in this way. The first level of processing raw texts represents a reliable morphological analysis - we give a survey of the effective implementation of the robust morphological analyser for Czech named ajka. Texts tagged by ajka can be further processed by the partial parser DIS and its extension VADIS which is based on verb valencies. The output of these systems serves for automatic partial disambiguation of input texts. The tools described in this paper are widely used for parsing large corpora and can be employed in the initial phase of semantic analysis. © Springer-Verlag Berlin Heidelberg 2003.