NeuSub: A Neural Submodular Approach for Citation Recommendation

Citation recommendation is a task that aims to automatically select suitable references for a working manuscript. This task has become increasingly urgent as the typical pools of candidates continue to grow, in the order of tens or hundreds of thousands or more. While several approaches to citation...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Binh Thanh Kieu, Inigo Jauregi Unanue, Son Bao Pham, Hieu Xuan Phan, Massimo Piccardi
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Citation recommendation deep neural networks structural/multiclass hinge loss submodular inference transformer models BERT Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/01ecc9f92f9244a58182820bc5f5c16a
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:01ecc9f92f9244a58182820bc5f5c16a
record_format	dspace
spelling	oai:doaj.org-article:01ecc9f92f9244a58182820bc5f5c16a2021-11-18T00:10:35ZNeuSub: A Neural Submodular Approach for Citation Recommendation2169-353610.1109/ACCESS.2021.3120727https://doaj.org/article/01ecc9f92f9244a58182820bc5f5c16a2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9576707/https://doaj.org/toc/2169-3536Citation recommendation is a task that aims to automatically select suitable references for a working manuscript. This task has become increasingly urgent as the typical pools of candidates continue to grow, in the order of tens or hundreds of thousands or more. While several approaches to citation recommendation have been proposed in the literature, they generally seem to lack principled mechanisms to ensure diversity and other global properties among the recommended citations. For this reason, in this paper we propose a novel citation recommendation approach that leverages a submodular scoring function and a deep document representation to achieve an effective trade-off between relevance to the query and diversity of the references. To optimally train the scoring function and the deep representation, we propose a novel training objective based on a structural/multiclass hinge loss and incremental recommendations. The experimental results over three popular citation datasets have showed that the proposed approach has led to remarkable accuracy improvements, with an increase of up to 1.91 pp of MRR and 3.29 pp of F1@100 score with respect to a state-of-the-art citation recommendation system.Binh Thanh KieuInigo Jauregi UnanueSon Bao PhamHieu Xuan PhanMassimo PiccardiIEEEarticleCitation recommendationdeep neural networksstructural/multiclass hinge losssubmodular inferencetransformer modelsBERTElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 148459-148468 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Citation recommendation deep neural networks structural/multiclass hinge loss submodular inference transformer models BERT Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Citation recommendation deep neural networks structural/multiclass hinge loss submodular inference transformer models BERT Electrical engineering. Electronics. Nuclear engineering TK1-9971 Binh Thanh Kieu Inigo Jauregi Unanue Son Bao Pham Hieu Xuan Phan Massimo Piccardi NeuSub: A Neural Submodular Approach for Citation Recommendation
description	Citation recommendation is a task that aims to automatically select suitable references for a working manuscript. This task has become increasingly urgent as the typical pools of candidates continue to grow, in the order of tens or hundreds of thousands or more. While several approaches to citation recommendation have been proposed in the literature, they generally seem to lack principled mechanisms to ensure diversity and other global properties among the recommended citations. For this reason, in this paper we propose a novel citation recommendation approach that leverages a submodular scoring function and a deep document representation to achieve an effective trade-off between relevance to the query and diversity of the references. To optimally train the scoring function and the deep representation, we propose a novel training objective based on a structural/multiclass hinge loss and incremental recommendations. The experimental results over three popular citation datasets have showed that the proposed approach has led to remarkable accuracy improvements, with an increase of up to 1.91 pp of MRR and 3.29 pp of F1@100 score with respect to a state-of-the-art citation recommendation system.
format	article
author	Binh Thanh Kieu Inigo Jauregi Unanue Son Bao Pham Hieu Xuan Phan Massimo Piccardi
author_facet	Binh Thanh Kieu Inigo Jauregi Unanue Son Bao Pham Hieu Xuan Phan Massimo Piccardi
author_sort	Binh Thanh Kieu
title	NeuSub: A Neural Submodular Approach for Citation Recommendation
title_short	NeuSub: A Neural Submodular Approach for Citation Recommendation
title_full	NeuSub: A Neural Submodular Approach for Citation Recommendation
title_fullStr	NeuSub: A Neural Submodular Approach for Citation Recommendation
title_full_unstemmed	NeuSub: A Neural Submodular Approach for Citation Recommendation
title_sort	neusub: a neural submodular approach for citation recommendation
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/01ecc9f92f9244a58182820bc5f5c16a
work_keys_str_mv	AT binhthanhkieu neusubaneuralsubmodularapproachforcitationrecommendation AT inigojauregiunanue neusubaneuralsubmodularapproachforcitationrecommendation AT sonbaopham neusubaneuralsubmodularapproachforcitationrecommendation AT hieuxuanphan neusubaneuralsubmodularapproachforcitationrecommendation AT massimopiccardi neusubaneuralsubmodularapproachforcitationrecommendation
_version_	1718425141302525952

NeuSub: A Neural Submodular Approach for Citation Recommendation

Ejemplares similares