Buch, Englisch, 116 Seiten, Format (B × H): 155 mm x 235 mm, Gewicht: 234 g
ISBN: 978-981-13-8936-8
Verlag: Springer Nature Singapore
This book describes recent advances in text summarization, identifies remaining gaps and challenges, and proposes ways to overcome them. It begins with one of the most frequently discussed topics in text summarization – ‘sentence extraction’ –, examines the effectiveness of current techniques in domain-specific text summarization, and proposes several improvements. In turn, the book describes the application of summarization in the legal and scientific domains, describing two new corpora that consist of more than 100 thousand court judgments and more than 20 thousand scientific articles, with the corresponding manually written summaries. The availability of these large-scale corpora opens up the possibility of using the now popular data-driven approaches based on deep learning. The book then highlights the effectiveness of neural sentence extraction approaches, which perform just as well as rule-based approaches, but without the need for any manual annotation. As a next step, multiple techniques for creating ensembles of sentence extractors – which deliver better and more robust summaries – are proposed. In closing, the book presents a neural network-based model for sentence compression. Overall the book takes readers on a journey that begins with simple sentence extraction and ends in abstractive summarization, while also covering key topics like ensemble techniques and domain-specific summarization, which have not been explored in detail prior to this.
Zielgruppe
Research
Autoren/Hrsg.
Fachgebiete
Weitere Infos & Material
1 Introduction
1.1 Extractive Summarization1.2 Information Fusion and Ensemble Techniques1.3 Abstractive Summarization1.4 Main contributions1.5 Organization
2 Related Work
2.1 Extractive Summarization2.1.1 Legal Document Summarization2.1.2 Scientific article Summarization2.2 Ensemble techniques for extractive summarization2.3 Sentence compression
3 Domain specific Extractive Summarization
3.1 Corpora3.2 Legal document Summarization3.2.1 Boosting legal vocabulary using a lexicon3.2.2 Weighted TextRank and LexRank3.2.3 Automatic key phrase identification3.2.4 Attention based sentence extractor3.3 Scientific article summarization3.4 Experiment Details3.4.1 Results3.5 Conclusion
4 Improving extractive techniques through rank aggregation
4.1 Motivation for rank aggregation4.2 Analysis of existing extractive systems4.2.1 Experimental Setup4.3 Ensemble of extractive summarization systems4.3.1 Effect of Informed fusion4.4 Discussion4.4.1 Determining the robustness of candidate systems4.4.2 Qualitative analysis of summaries
5 Leveraging content similarity in summaries for generating better ensembles
5.1 Limitations of consensus based aggregation5.2 Proposed approach for content based aggregation5.3 Document level aggregation5.3.1 Experimental results5.4 Sentence Level aggregation5.4.1 SentRank5.4.2 GlobalRank5.4.3 LocalRank5.4.4 HybridRank5.4.5 Experimental Results5.5 Conclusion
6 Neural model for sentence compression
6.1 Sentence compression by deletion6.2 Sentence compression using Sequence to Sequence model6.2.1 Sentence Encoder6.2.2 Context Encoder6.2.3 Decoder6.2.4 Attention module6.3 Exploiting SMT techniques for sentence compression6.4 Results for sentence compression6.5 Limitations of sentence compression techniques6.6 Overall System
7 Conclusion and Future Work