Discourse segmentation in aid of document summarization

Boguraev, Branimir; Neff, Mary S.

doi:10.1109/hicss.2000.926687

Cited by 54 publications

(52 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Hearst's algorithm uses a sliding window to compute block similarity at regular intervals in the text and then determines subtopic boundaries by looking for significant dips in the resulting similarity plot. Variations of this method have been employed by Richmond et al [1997], Salton et al [1996], and Boguraev and Neff [2000]. Reynar [1998] extends Hearst's [1997] ideas by representing term overlap in a matrix and then applying an optimization algorithm to determine boundaries.…”

Section: Related Workmentioning

confidence: 99%

Broad coverage paragraph segmentation across languages and domains

Sporleder

Lapata

2006

ACM Trans. Speech Lang. Process.

View full text Add to dashboard Cite

This article considers the problem of automatic paragraph segmentation. The task is relevant for speech-to-text applications whose output transcipts do not usually contain punctuation or paragraph indentation and are naturally difficult to read and process. Text-to-text generation applications (e.g., summarization) could also benefit from an automatic paragaraph segementation mechanism which indicates topic shifts and provides visual targets to the reader. We present a paragraph segmentation model which exploits a variety of knowledge sources (including textual cues, syntactic and discourse-related information) and evaluate its performance in different languages and domains. Our experiments demonstrate that the proposed approach significantly outperforms our baselines and in many cases comes to within a few percent of human performance. Finally, we integrate our method with a single document summarizer and show that it is useful for structuring the output of automatically generated text.

show abstract

Section: Related Workmentioning

confidence: 99%

Broad coverage paragraph segmentation across languages and domains

Sporleder

Lapata

2006

ACM Trans. Speech Lang. Process.

View full text Add to dashboard Cite

show abstract

“…Lam et al [99] use an existing IBM summarization system [17] to produce summaries. However, in order to make this feasible, they first must pre-process the message to remove pieces that would be problematic; their pre-processor removes certain headers, quoted text, forward information and email signatures.…”

Section: Summarization Of Individual Email Messagesmentioning

confidence: 99%

Automatic Summarization

Nenkova

2011

FNT in Information Retrieval

333

212

View full text Add to dashboard Cite

“…In order to identify the window, lexical repetition is used based on [26] who pointed out that cohesion can best be explained by focusing on how lexical repetition is manifested, in numerous ways, across pairs of sentences or utterances spoken by criminals through the telephone or other forms of communication. In order to identify the similarity sim between two utterances, i U and j U , cosine similarity measure is used in which the cosine of i U…”

Section: Text Analysismentioning

confidence: 99%

Duo-mining with multi-agents: the next wave to control organised crime

Chibelushi

Bentley²

2008

WIT Transactions on Information and Communication Technologies

View full text Add to dashboard Cite

Text and data mining are fast growing areas and are believed to have high commercial potential value in knowledge discovery and information filtering areas of application. Although text mining manages unstructured data, most of knowledge discovery and information filtering can be done using data mining. Despite that, both technologies do not actively predict and prevent problems, instead they leave the work to the experts to manually interpret the data, anticipate future events and make the final decision. This paper proposes the ASKARI approach outlined in this paper, which combines duo-mining (text and data mining) with multi-agent systems, the approach aims to predict and prevent crimes before they happen, and in fact, might become the next wave of knowledge discovery. Also, the paper highlights the benefits of combining duomining and multi-agents in prediction and preventing organised crimes.

show abstract

Discourse segmentation in aid of document summarization

Cited by 54 publications

References 19 publications

Broad coverage paragraph segmentation across languages and domains

Broad coverage paragraph segmentation across languages and domains

Automatic Summarization

Duo-mining with multi-agents: the next wave to control organised crime

Contact Info

Product

Resources

About