In this paper, we describe a method of automatically generating summary slides from a text. The slides are generated by itemizing topic/non-topic parts that are extracted from the text based on syntactic/case analysis. The indentations of the items are controlled according to the discourse structure, which is detected by cue phrases, identification of word chain and similarity between two sente...

Traditional approaches for extractive summarization score/classify sentences based on features such as position in the text, word frequency and cue phrases. These features tend to produce satisfactory summaries, but have the inconvenience of being domain dependent. In this paper, we propose to tackle this problem representing the sentences by word sequences (n-grams), a widely used representati...

We propose a novel method to predict the interparagraph discourse structure of text, i.e. to infer which paragraphs are related to each other and form larger segments on a higher level. Our method combines a clustering algorithm with a model of segment “relatedness” acquired in a machine learning step. The model integrates information from a variety of sources, such as word co-occurrence, lexic...

We present an unsupervised approach to recognizing discourse relations of CONTRAST, EXPLANATION-EVIDENCE, CONDITION and ELABORATION that hold between arbitrary spans of texts. We show that discourse relation classifiers trained on examples that are automatically extracted from massive amounts of text can be used to distinguish between some of these relations with accuracies as high as 93%, even...

This paper presents research that connects the cohesion structure of a text to the derivation of its coherence structure. Two diierent algorithms that derive the cohesion structure in the form of lexical paths from large thesauri are illustrated. Their results are correlated with (1) cue phrases of discourse usage and (2) coherence constraints empirically derived. A novel model of coherence str...

We have developed a set of microplanning choice rules which are intended to enable Natural Language Generation (NLG) systems to generate appropriate texts for readers with below-average literacy, focusing in particular on choices related to how discourse structure is expressed (cue phrases, ordering, sentence structure). Evaluation experiments suggest that our rules do enhance the readability o...

• Cue-based retrieval cannot explain encoding interference in sentence processing. An extension to cue-based can do so, and it makes reading time predictions. A self-organization-based model diverging Two self-paced experiments provide support for self-organization. According theories of comprehension, establishing the syntactic dependency between a verb grammatical subject is susceptible from ...

Coherence relations have usually been taken to link clauses and larger units. After arguing that some phrases can be seen as discourse units, a computational account for such phrases is presented that integrates surface-based criteria with inferential ones. This approach can be generalized to treat intra-sentential cue-phrases. Since cue-phrases are not always present, referential relations bet...

The influence of sentential cues (such as animacy and word order) on thematic role interpretation differs a function language (MacWhinney et al. 1984). However, existing cross-linguistic research has typically focused transitive sentences involving agents, non-default verb classes is less well understood. Here, we compared the way in which English German native speakers – languages known to dif...

We outline how utterances in dialogs can be interpreted using a partial rst order logic. We exploit the capability of this logic to talk about the truth status of formulae to deene a notion of coherence between utterances and explain how this coherence relation can serve for the construction of AND/OR trees that represent the segmentation of the dialog. In a BDI model we formalize basic assumpt...