Topic segmentation plays an important role for discourse analysis and document understanding. Previous work mainly focus on unsupervised method for topic segmentation. In this paper, we propose to use bidirectional long shortterm memory(BLSTM) model, along with convolutional neural network(CNN) for learning paragraph representation. Besides, we present a novel algorithm based on frequent subseq...

We propose a method for automatically identifying rhetorical relations. We use supervised machine learning but exploit cue phrases to automatically extract and label training data. Our models draw on a variety of linguistic cues to distinguish between the relations. We show that these feature-rich models outperform the previously suggested bigram models by more than 20%, at least for small trai...

Both rhetorical structure and punctuation have been helpful in discourse processing. Based on a corpus annotation project, this paper reports the discursive usage of 6 Chinese punctuation marks in news commentary texts: Colon, Dash, Ellipsis, Exclamation Mark, Question Mark, and Semicolon. The rhetorical patterns of these marks are compared against patterns around cue phrases in general. Result...

The problem of information overload is hot issue with growth of worldwide web. The need for the tools those should be able to absorb this huge information and eliminate this problem is evident especially for IR systems. Text is not a simple sequence of words but carries a structure. It is essential to handle these uncontrollable complex structures of sentence, grammatical and lexical irrelevanc...

Expectations about the correlation of cue phrases, the duration of unfilled pauses and the structuring of spoken discourse are framed in light of Grosz and Sidner’s theory of discourse and are tested for a directions-giving dialogue. The results suggest that cue phrase and discourse structuring tasks may align, and show a correlation for pause length and some of the modifications that speakers ...

The present study explores the vocal intensity of turn-initial cue phrases in a corpus of dialogues in Swedish. Cue phrases convey relatively little propositional content, but have several important pragmatic functions. The majority of these entities are frequently occurring monosyllabic words such as “eh”, “mm”, “ja”. Prosodic analysis shows that these words are produced with higher intensity ...

This paper presents an approach to automatic segmentation of text written in English into Elementary Discourse Units (EDUs) using syntactic information and cue phrases. The system takes documents with syntactic information as the input and generates EDUs as well as their nucleus/satellite roles. The experiment shows that this approach gives promising results in comparison with some of the promi...

The most important part of patent specification is where the claims are written. It is common that claims written in Japanese are described in one sentence with peculiar style and are difficult to understand for ordinary people. We are investigating NLP technologies to improve readability of patent claims. To do so, it is necessary to present the structure of patent claims in a readable way. We...

This paper describes experiments to extract discourse relations holding between two text spans in Swedish. We considered three relation types: cause-explanation-evidence (CEV), contrast, and elaboration and we extracted word pairs eliciting these relations. We determined a list of Swedish cue phrases marking explicitly the relations and we learned the word pairs automatically from a corpus of 6...

Scientific citations play a crucial role in maintaining the network of relationships among mutually relevant articles within a research field. Customarily, authors include citations in their papers to indicate works that are foundational in their field, background for their own work, or representative of complementary or contradictory research. But, determining the nature of the exact relations...