Using error decay prediction to overcome practical issues of deep active learning for named entity recognition

DOI10.1007/S10994-020-05897-1MaRDI QIDQ2217437zbMATH OpenOpenAlexFDO

Authors Haw-Shiuan Chang, Shankar Vembu, Sunil Mohan, Rheeya Uppaal, Andrew Mccallum

Publication date 29 December 2020

Published in Machine Learning (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1911.07335

zbMATH Keywords

active learningblack-box modelsclusteringnamed entity recognitionrobustness to labeling noisetransparency

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• A method based on multi-standard active learning to recognize entities in electronic medical record
• scientific article; zbMATH DE number 7113026
• Adaptive active learning for semi-supervised learning
• Active learning for cost-sensitive classification
• Evidence-based uncertainty sampling for active learning

Cites work

• scientific article; zbMATH DE number 3340881 (Why is no real title available?)
• Natural language processing (almost) from scratch
• Newton-Type Minimization via the Lanczos Method
• On the hardness of approximating minimization problems
• Probability and computing. Randomization and probabilistic techniques in algorithms and data analysis
• Two faces of active learning

Cited in

(1)

• scientific article; zbMATH DE number 7113026 (Why is no real title available?)

Describes a project that uses

Uses Software

• word2vec
• tn

This page was built for publication: Using error decay prediction to overcome practical issues of deep active learning for named entity recognition

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2217437)