Method and dataset entity mining in scientific literature: A CNN + BiLSTM model with self-attention

The traditional literature analysis mainly focuses on the literature metadata such as topics, authors, keywords, references, and rarely pays attention to the main content of papers. However, in many scientific domains such as science, computing, engineering, the methods and datasets involved in the...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Knowledge-based systems 2022-01, Vol.235, p.107621, Article 107621
Hauptverfasser:	Hou, Linlin, Zhang, Ji, Wu, Ou, Yu, Ting, Wang, Zhen, Li, Zhao, Gao, Jianliang, Ye, Yingchun, Yao, Rujing
Format:	Artikel
Sprache:	eng
Schlagworte:	Ablation Artificial neural networks CNN+BiLSTM-Attention-CRF structure Datasets Domains Literature analysis Methods and datasets mining Named entity recognition Recommender systems Scientific papers
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!