Annotating discourse connectives by looking at their translation: The translation-spotting technique

The various meanings of discourse connectives like while and however are difficult to identify and annotate, even for trained human annotators. This problem is all the more important that connectives are salient textual markers of cohesion and need to be correctly interpreted for many NLP applications. In this paper, we suggest an alternative route to reach a reliable annotation of connectives, by making use of the information provided by their translation in large parallel corpora. This method thus replaces the difficult explicit reasoning involved in traditional sense annotation by an empirical clustering of the senses emerging from the translations. We argue that this method has the advantage of providing more reliable reference data than traditional sense annotation. In addition, its simplicity allows for the rapid constitution of large annotated datasets.

Automatic identification of discourse markers in dialogues: An in-depth study of like and well

The lexical items like and well can serve as discourse markers (DMs), but can also play numerous other roles, such as verb or adverb. Identifying the occurrences that function as DMs is an important step for language understanding by computers. In this study, automatic classifiers using lexical, prosodic/positional and sociolinguistic features are trained over transcribed dialogues, manually annotated with DM information. The resulting classifiers improve state-of-the-art performance of DM identification, at about 90% recall and 79% precision for like (84.5% accuracy, κ = 0.69), and 99% recall and 98% precision for well (97.5% accuracy, κ = 0.88). Automatic feature analysis shows that lexical collocations are the most reliable indicators, followed by prosodic/positional features, while sociolinguistic features are marginally useful for the identification of DM like and not useful for well. The differentiated processing of each type of DM improves classification accuracy, suggesting that these types should be treated individually.

Annotating the meaning of discourse connectives in multilingual corpora

Discourse connectives are lexical items indicating coherence relations between discourse segments. Even though many languages possess a whole range of connectives, important divergences exist cross-linguistically in the number of connectives that are used to express a given relation. For this reason, connectives are not easily paired with a univocal translation equivalent across languages. This paper is a first attempt to design a reliable method to annotate the meaning of discourse connectives cross-linguistically using corpus data. We present the methodological choices made to reach this aim and report three annotation experiments using the framework of the Penn Discourse Tree Bank.

Discourse-level Annotation over Europarl for Machine Translation: Connectives and Pronouns

This paper describes methods and results for the annotation of two discourse-level phenomena, connectives and pronouns, over a multilingual parallel corpus. Excerpts from Europarl in English and French have been annotated with disambiguation information for connectives and pronouns, for about 3600 tokens. This data is then used in several ways: for cross-linguistic studies, for training automatic disambiguation software, and ultimately for training and testing discourse-aware statistical machine translation systems. The paper presents the annotation procedures and their results in detail, and overviews the first systems trained on the annotated resources and their use for machine translation.

Towards Automatic Identification of Discourse Markers in Dialogs: the Case of 'Like'

This article discusses the detection of discourse markers (DM) in dialog transcriptions, by human annotators and by automated means. After a theoretical discussion of the definition of DMs and their relevance to natural language processing, we focus on the role of like as a DM. Results from experiments with human annotators show that detection of DMs is a difficult but reliable task, which requires prosodic information from soundtracks. Then, several types of features are defined for automatic disambiguation of like: collocations, part-of-speech tags and duration-based features. Decision-tree learning shows that for like, nearly 70% precision can be reached, with near 100% recall, mainly using collocation filters. Similar results hold for well, with about 91% precision at 100% recall.