In this paper we investigate the benefit of stochastic predictor components for the parsing quality which can be obtained with a rule-based dependency grammar. By including a chunker, a supertagger, a PP attacher, and a fast probabilistic parser we were able to improve upon the baseline by , bringing the overall labelled accuracy to on the German NEGRA corpus. We attribute the successful integration to the ability of the underlying grammar model to combine uncertain evidence in a soft manner, thus avoiding the problem of error propagation. .
