Reverse-engineering language: A study on the semantic compositionality of German compounds

Abstract

In this paper we analyze the performance of different composition models on a large dataset of German compound nouns. Given a vector space model for the German language, we try to reconstruct the observed representation (the corpusestimated vector) of a compound by composing the observed representations of its two immediate constituents. We explore the composition models proposed in the literature and also present a new, simple model that achieves the best performance on our dataset..