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
07 Nov 2022
TL;DR: In this paper , an integrated approach for text summarization using images, table labels, etc. is presented, where the first step is text processing and the second step involves using graph-based algorithms to extract the most important sentences from the document.
Abstract: Analyzing text and image data is always time-consuming, and with the rapid growth in the amount of data, important meanings of the information may be lost. Natural language processing (NLP) and graph-based methods can be used to summarize large documents, The system proposed in this chapter is an integrated approach for text summarization using images, table labels, etc. The first step is text processing, which extracts important text information. The second step involves using graph-based algorithms to extract the most important sentences from the document. Techniques such as node ranking aim to extract the meaningful information from the document. A machine learning (ML) model is used to train the system and provide an accurate document summary. NLP-based text summarization opens up various applications areas, such as e-learning, meeting summarization, e-news systems, social network data analysis, rapid decision support in business analysis, and many more.
04 Dec 2014
TL;DR: This work proposes a mechanism that can overcome difficulties and generate update summaries in an affordable way because of the inability to keep summaries unaffected by the order in which new documents are added to the pool and the need to process whole set of documents each time the summary needs to be updated.
Abstract: Documents which are published both online and offline are considered to be the primary source of information. Astonishing growth of documentation and communication systems tends to flood these pools of information sources with enormous amount of documents. In such a scenario, it is critical to envisage algorithms and methodologies that can convert these huge collection of documents to their best possible form of summaries. It will help to cater to the information hunters who needs only the abstract summaries in a fully digestible form. Different methods which can perform this task can be compared on the basis of quality of summary that it generates and the amount of processing power that it demands. Existing methods are capable of generating summaries incrementally (update summaries as and when new documents are added to the pool). But the inability to keep summaries unaffected by the order in which new documents are added to the pool and the need to process whole set of documents (together with those which are already summarized)each time the summary needs to be updated, pulls them back from their potential applications. We propose a mechanism that can overcome these difficulties and generate update summaries in an affordable way.
Dissertation•
30 Jun 2011
TL;DR: The development of an automatic summarization system which draws on the conceptual idea of the Pyramid evaluation scheme and the techniques developed for the proposed evaluation system, and the development of a fully automated evaluation method.
Abstract: Summarization is the process of creating a more compact textual representation of a document or a collection of documents. In view of the vast increase in electronically available information sources in the last decade, filters such as automatically generated summaries are becoming ever more important to facilitate the efficient acquisition and use of required information. Different methods using natural language processing (NLP) techniques are being used to this end. One of the shallowest approaches is the clustering of available documents and the representation of the resulting clusters by one of the documents; an example of this approach is the Google News website. It is also possible to augment the clustering of documents with a summarization process, which would result in a more balanced representation of the information in the cluster, NewsBlaster being an example. However, while some systems are already available on the web, summarization is still considered a difficult problem in the NLP community. One of the major problems hampering the development of proficient summarization systems is the evaluation of the (true) quality of system-generated summaries. This is exemplified by the fact that the current state-of-the-art evaluation method to assess the information content of summaries, the Pyramid evaluation scheme, is a manual procedure. In this light, this thesis has three main objectives. 1. The development of a fully automated evaluation method. The proposed scheme is rooted in the ideas underlying the Pyramid evaluation scheme and makes use of deep syntactic information and lexical semantics. Its performance improves notably on previous automated evaluation methods. 2. The development of an automatic summarization system which draws on the conceptual idea of the Pyramid evaluation scheme and the techniques developed for the proposed evaluation system. The approach features the algorithm for determining the pyramid and bases importance on the number of occurrences of the variable-sized contributors of the pyramid as opposed to word-based methods exploited elsewhere. 3. The development of a text coherence component that can be used for obtaining the best ordering of the sentences in a summary.
07 Dec 2018
TL;DR: This work states that with the recent success of sequence-to-sequence techniques, which are based on recurrent neural networks, there has been an increase in the amount of attention paid to automatic abstractive summarization techniques.
Abstract: ive summarization, where new words and sentences that may not have been present in the original document or collection of documents are generated in the summary, is a difficult task, because it requires an understanding of the contents of the documents. As such, more research has historically been done in extractive summarization techniques than in abstractive ones. However, with the recent success of sequence-to-sequence techniques, which are based on recurrent neural networks, there has been an increase in the amount of attention paid to automatic abstractive summarization techniques. [2]
01 Jan 2015
TL;DR: This paper is analysing the way to automatic generation of the presentation slides from academic papers and proposes PPSGen system, which have lots of advantages over baseline methods.
Abstract: Presentation-based learning is an effective way of learning in which students or employee of the organization are able to receive continuous feedback from teammates or from their coaches. It is the pictorial way to represent the work. Before going towards the presentation of the work presenters have to work on the slides. These slides of presentation are made from the article, some academic papers or with the help of internet. It results into wastage of more timing to create slides rather than focusing on preparation of the presentation. In this paper, we are analysing the way to automatic generation of the presentation slides from academic papers. Due to this the presenters can prepare their formal slides in a quicker way. Therefore, we are proposing PPSGen system to address this problem of existing system. PPSGen have lots of advantages over baseline methods.