Recent work on bridging resolution has so far been based on the corpus ISNotes (Markert et al. 2012), as this was the only corpus available with unrestricted bridging annotation. Hou et al. 2014’s rule-based system currently achieves state-of-the-art performance on this corpus, as learning-based approaches suffer from the lack of available training data. Recently, a number of new corpora with bridging annotations have become available. To test the generalisability of the approach by Hou et al. 2014, we apply a slightly extended rule-based system to these corpora. Besides the expected out-of-domain effects, we also observe low performance on some of the in-domain corpora. Our analysis shows that this is the result of two very different phenomena being defined as bridging, namely referential and lexical bridging. We also report that filtering out gold or predicted coreferent anaphors before applying the bridging resolution system helps improve bridging resolution.

Adding manually annotated prosodic information, specifically pitch accents and phrasing, to the typical text-based feature set for coreference resolution has previously been shown to have a positive effect on German data. Practical applications on spoken language, however, would rely on automatically predicted prosodic information. In this paper we predict pitch accents (and phrase boundaries) using a convolutional neural network (CNN) model from acoustic features extracted from the speech signal. After an assessment of the quality of these automatic prosodic annotations, we show that they also significantly improve coreference resolution.

DIRNDL is a spoken and written corpus based on German radio news, which features coreference and information-status annotation (including bridging anaphora and their antecedents), as well as prosodic information. We have recently extended DIRNDL with a fine-grained two-dimensional information status labeling scheme. We have also applied a state-of-the-art part-of-speech and morphology tagger to the corpus, as well as highly accurate constituency and dependency parsers. In the light of this development we believe that DIRNDL is an interesting resource for NLP researchers working on automatic coreference and bridging resolution. In order to enable and promote usage of the data, we make it available for download in an accessible tabular format, compatible with the formats used in the CoNLL and SemEval shared tasks on automatic coreference resolution.

We provide a robust and detailed annotation scheme for information status, which is easy to use, follows a semantic rather than cognitive motivation, and achieves reasonable inter-annotator scores. Our annotation scheme is based on two main assumptions: firstly, that information status strongly depends on (in)definiteness, and secondly, that it ought to be understood as a property of referents rather than words. Therefore, our scheme banks on overt (in)definiteness marking and provides different categories for each class. Definites are grouped according to the information source by which the referent is identified. A special aspect of the scheme is that non-anaphoric expressions (e.g.\ names) are classified as to whether their referents are likely to be known or unknown to an expected audience. The annotation scheme provides a solution for annotating complex nominal expressions which may recursively contain embedded expressions. In annotating a corpus of German radio news bulletins, a kappa score of .66 for the full scheme was achieved, a core scheme of six top-level categories yields kappa = .78.