Multi-label legal document classification: A deep learning-based approach with label-attention and domain-specific pre-training

Song, Dezhao; Vold, Andrew; Madan, Kanika; Schilder, Frank

doi:10.1016/j.is.2021.101718

Cited by 44 publications

(44 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Em comparação com os trabalhos apresentados na Seção 5, que fazem uso do BERT e do ULMFiT, é preciso considerar que a maioria realiza a classificação multilabel em um dataset com pelo menos • 7 66 classes, sendo que os trabalhos [Song et al 2021], [Chalkidis et al 2020] e [Shaheen et al 2020…”

Section: Resultados Na Tarefa De Classificação Textualunclassified

“…(Q.2) Dos sete trabalhos estudados nessa categoria, verifica-se que a maioria ( [Song et al 2021], [Wang et al 2020], [Chalkidis et al 2020], [Shaheen et al 2020] e [Chalkidis et al 2019]) utiliza o modelo BERT, e os outros dois [Soh et al 2019] e [Campos et al 2020], tem como principal modelo o ULMFiT, todos alcançando F1-score acima de 0.8. Dentre os trabalhos, apenas [Wang et al 2020] e [Campos et al 2020] são em língua portuguesa do Brasil, os demais em língua inglesa, e um deles [Shaheen et al 2020], apresenta um modelo multilingual que, além do inglês, inclui o francês e o alemão.…”

Section: Classificação Textual No Domínio Jurídico Com Modelos De Linguagensunclassified

See 1 more Smart Citation

Text Classification in Law Area: a Systematic Review

Martins¹,

Silva²

2021

Anais Do IX Symposium on Knowledge Discovery, Mining and Learning (KDMiLe 2021)

View full text Add to dashboard Cite

Automatic Text Classification represents a great improvement in law area workflow, mainly in the migration of physical to electronic lawsuits. A systematic review of studies on text classification in law area from January 2017 up to February 2020 was conducted. The search strategy identified 20 studies, that were analyzed and compared. The review investigates from research questions: what are the state-of-art language models, its application of text classification in English and Brazilian Portuguese datasets from legal area, if there are available language models trained on Brazilian Portuguese, and datasets in Brazilian law area. It concludes that there are applications of automatic text classification in Brazil, although there is a gap on the use of language models when compared with English language dataset studies, also the importance of language model in domain pre-training to improve results, as well as there are two studies making available Brazilian Portuguese language models, and one introducing a dataset in Brazilian law area.

show abstract

Section: Resultados Na Tarefa De Classificação Textualunclassified

Section: Classificação Textual No Domínio Jurídico Com Modelos De Linguagensunclassified

Text Classification in Law Area: a Systematic Review

Martins¹,

Silva²

2021

Anais Do IX Symposium on Knowledge Discovery, Mining and Learning (KDMiLe 2021)

View full text Add to dashboard Cite

show abstract

“…The Chinese text, which is never evenly distributed and hence Long Short Memory, Convolution Neural Network and attention algorithm model with weight for key features are used for classification of Chinese word [30]. Legal documents are classified based on Multi-label text classification [31]. The cloud dataset is classified based on image patches available on the cloud dataset [32].…”

Section: Related Workmentioning

confidence: 99%

SF-CNN: Deep Text Classification and Retrieval for Text Documents

Sarasu¹,

Thyagharajan²,

Shanker³

2023

Intelligent Automation &Amp; Soft Computing

View full text Add to dashboard Cite

“…It has been applied successfully to various computer vision tasks such as food image recognition, 35 mammogram image classication 36 and multi-label legal document classication. 37 Carbendazim is a type of broad-spectrum benzimidazole fungicide, which has been commonly employed to control plant diseases in cereals and fruits. Rice is an important food crop for human beings, which a wide area has been cultivated for.…”

Section: Introductionmentioning

confidence: 99%

“…It has been applied successfully to various computer vision tasks such as food image recognition, 35 mammogram image classification 36 and multi-label legal document classification. 37 …”

Section: Introductionmentioning

confidence: 99%