Automatic related work section generation: experiments in scientific document abstracting
Mostra el registre complet Registre parcial de l'ítem
- dc.contributor.author AbuRa'ed, Ahmed Ghassan Tawfiq
- dc.contributor.author Saggion, Horacio
- dc.contributor.author Shvets, Alexander
- dc.contributor.author Bravo Serrano, Àlex, 1984-
- dc.date.accessioned 2020-07-30T06:44:31Z
- dc.date.issued 2020
- dc.description.abstract Related work sections or literature reviews are an essential part of every scientific article being crucial for paper reviewing and assessment. However, writing a good related work section is an activity which requires considerable expertise to identify, condense/summarize, and combine relevant information from different sources. In this work we compare different automatic methods to produce “descriptive” related work sections given as input the set of papers which need to be described. The main contribution of our work is a neural sequence learning process which produces citation sentences to be included in a related work section of an article. We train the neural architecture using an available scientific data set of citation sentences and we test over a data set of related work sections; we also compare the performance to a set of baseline extractive summarizers, an abstractive summarizer and a state of the art CNNs approach. Our results indicate that our approach outperforms the simple as well as the informed baselines.
- dc.description.sponsorship This work is (partly) supported by the Spanish Ministry of Economy and Competitiveness under the Maria de Maeztu Units of Excellence Programme (MDM-2015-0502) Special thanks to ALI TAQI for designing the pointer generator architecture figure.
- dc.format.mimetype application/pdf
- dc.identifier.citation AbuRa’ed A, Saggion H, Shvets A, Bravo A. Automatic related work section generation: experiments in scientific document abstracting. Scientometrics. 2020;125(3):3159-85. DOI: 10.1007/s11192-020-03630-2
- dc.identifier.doi http://dx.doi.org/10.1007/s11192-020-03630-2
- dc.identifier.issn 0138-9130
- dc.identifier.uri http://hdl.handle.net/10230/45219
- dc.language.iso eng
- dc.publisher Springer
- dc.relation.ispartof Scientometrics. 2020;125(3):3159-85
- dc.rights © Springer The final publication is available at Springer via http://dx.doi.org/10.1007/s11192-020-03630-2
- dc.rights.accessRights info:eu-repo/semantics/openAccess
- dc.subject.keyword Scientific summarization
- dc.subject.keyword Document abstracting
- dc.subject.keyword Sequence learning
- dc.subject.keyword Information extraction from scientific literature
- dc.title Automatic related work section generation: experiments in scientific document abstracting
- dc.type info:eu-repo/semantics/article
- dc.type.version info:eu-repo/semantics/acceptedVersion