No abstract

Making use of latent semantic analysis, we explore the hypothesis that local linguistic context can serve to identify multi-word expressions that have noncompositional meanings. We propose that vector-similarity between distribution vectors associated with an MWE as a whole and those associated with its constitutent parts can serve as a good measure of the degree to which the MWE is compositional. We present experiments that show that low (cosine) similarity does, in fact, correlate with non-compositionality.

This paper presents evidence from both corpus studies and psycholinguistic experiments regarding the referential preferences of German personal pronouns (er sie, es) and demonstrative pronouns (der, die, das). Both types of pronouns are naturally used in German to refer to both animate and inanimate referents, but they appear to have different preferences with respect to their choice of antecedents: While personal pronouns show a slight preference for antecedents which are grammatical subjects, demonstrative pronouns show a strong bias for nonsubject antecedents.

No abstract

The aim of this paper is to present a language-neutral, theory-neutral method for annotating sentenceinternal temporal relations. The annotation method is simple and can be applied without special training. The annotations are provided with a well-defined model-theoretic interpretation for use in the content-based comparison of annotations. Temporally annotated corpora have a number of applications, in lexicon/induction, translation and linguistic investigation. A searchable multi-language database has already been created.