Topic
Multi-document summarization
About: Multi-document summarization is a research topic. Over the lifetime, 2270 publications have been published within this topic receiving 71850 citations.
Papers published on a yearly basis
Papers
More filters
01 Jan 2014
TL;DR: This work proposes a two-step approach towards domain independent generic video summarization incorporating video categorization for enhanced keyframe extraction and talks about building effective mechanisms for large scale categorization over big hierarchical category tree of videos.
Abstract: Over the past few years, there has been a massive increase in amount of video content created. Massive growth in video content poses problem of information overload and management of content. In order to manage the growing videos on the web and also to extract efficient and valid information from the videos, more attention has to be paid towards video and image processing technologies. Video summaries provide condensed and succinct representations of the content of a video stream through a combination of still images, video segments, graphical representations and textual descriptors. Existing video summarization techniques have attempted to solve the problem of condensing the content of a video in domain specific manner. However, such domain specific summarization mechanisms do not generalize well over different genres of videos. To make video summarization scalable enough to cater to the needs of growing massive online video content, it needs to be generic and adaptable for its applicability on any category of video. This work presents a general approach towards video summarization process and proposes a two-step approach towards domain independent generic video summarization incorporating video categorization for enhanced keyframe extraction. This paper also talks about building effective mechanisms for large scale categorization over big hierarchical category tree of videos.
1 citations
Journal Article•
TL;DR: The proposed Entity-Description-Utility model, a general expandable model, starts from video entity, gets utilities after descriptions, and finally generates video summarization by the utilities.
Abstract: The current video summarization techniques are in want of a normal and expandable summarization model.To solve this problem,Entity-Description-Utility(EDU),a general expandable model,is proposed.The model starts from video entity,gets utilities after descriptions,and finally generates video summarization by the utilities.The EDU model is described in detail, and a method of news story summarization based on this model is also produced.The experiment proves the effectiveness of the method.
1 citations
Journal Article•
TL;DR: According to the cogent experimental results, the very application of automatic text summarization is not restrained by the different avenues of inquiry, and can effectively delete out the prolixity and better demonstrate the textual contents.
Abstract: With the rapid development of information technology,the information on Internet has been greatly increased.The automatic text summarization technology is focused on at present as a hot issue.This paper puts forward a new method of automatic text summarization based on vector space model,which automatically processes prolixity.This approach automatically categorizes the statistics and sorts them out,and then by applying vector space model,further processes the prolixity of the text summary.According to the cogent experimental results,the very application of automatic text summarization is not restrained by the different avenues of inquiry,on the contrary,it can effectively delete out the prolixity and better demonstrate the textual contents.
1 citations
Posted Content•
TL;DR: In this paper, an extractive summarizer for multi-document setting is proposed. But, the authors focus on the coherence of summary coherence for increasing readability of the summary.
Abstract: In this work, we aim at developing an extractive summarizer in the multi-document setting. We implement a rank based sentence selection using continuous vector representations along with key-phrases. Furthermore, we propose a model to tackle summary coherence for increasing readability. We conduct experiments on the Document Understanding Conference (DUC) 2004 datasets using ROUGE toolkit. Our experiments demonstrate that the methods bring significant improvements over the state of the art methods in terms of informativity and coherence.
1 citations
TL;DR: This study proposes a template-based method of automatic summarization of multiple news articles using the semantic categories of sentences, where the template filled with simple sentences rather than original long sentences is used to generate a summary for an event/accident.
Abstract: This study proposes a template-based method of automatic summarization of multiple news articles using the semantic categories of sentences. First, the semantic categories for core information to be included in a summary are identified from training set of documents and their summaries. Then, cue words for each slot of the template are selected for later classification of news sentences into relevant slots. When a news article is input, its event/accident category is identified, and key sentences are extracted from the news article and filled in the relevant slots. The template filled with simple sentences rather than original long sentences is used to generate a summary for an event/accident. In the user evaluation of the generated summaries, the results showed the 54.l% recall ratio and the 58.l% precision ratio in essential information extraction and 11.6% redundancy ratio.