Deep Active Learning for Named Entity Recognition

Shen, Yanyao; Yun, Hyokun; Lipton, Zachary C.; Kronrod, Yakov; Anandkumar, Animashree

doi:10.18653/v1/w17-2630

Cited by 308 publications

(280 citation statements)

References 28 publications

Supporting

Mentioning

275

Contrasting

Unclassified

Order By: Relevance

“…The dataset contains several sources of noise, including OCR errors, referencing errors or inconsistencies, annotation errors. In part for this reason, we consider as the most important next step for future work to explore how active learning or semi-supervised learning techniques might be used in order to maximize the model gain while at the same time minimizing the costly process of manual annotation (Peters et al, 2017;Shen et al, 2018). At the same time, we plan to explore how to align and use existing annotated datasets with coverage in the arts and humanities (Anzaroot and McCallum, 2013).…”

Section: Resultsmentioning

confidence: 99%

Deep Reference Mining From Scholarly Literature in the Arts and Humanities

Alves

Colavizza

Kaplan

2018

Front. Res. Metr. Anal.

View full text Add to dashboard Cite

We consider the task of reference mining: the detection, extraction and classification of references within the full text of scholarly publications. Reference mining brings forward specific challenges, such as the need to capture the morphology of highly abbreviated words and the dependence among the elements of a reference, both following codified reference styles. This task is particularly difficult, and little explored, with respect to the literature in the arts and humanities, where references are mostly given in footnotes. We apply a deep learning architecture for reference mining from the full text of scholarly publications. We explore and discuss three architectural components: word and character-level word embeddings, different prediction layers (Softmax and Conditional Random Fields) and multi-task over single-task learning. Our best model uses both pre-trained word embeddings and characters embeddings, and a BiLSTM-CRF architecture. We test our solution on a dataset of annotated references from the historiography on Venice and, using a linear-chain CRF classifier as a baseline, we show that this deep learning architecture improves by a considerable margin. Furthermore, multi-task learning performs almost on par with a single-task approach. We thus confirm that there are important gains to be had by adopting deep learning for the task of reference mining.

show abstract

Section: Resultsmentioning

confidence: 99%

Deep Reference Mining From Scholarly Literature in the Arts and Humanities

Alves

Colavizza

Kaplan

2018

Front. Res. Metr. Anal.

View full text Add to dashboard Cite

show abstract

“…To the best of our knowledge, active learning has not been applied to classification tasks for scientific text yet. Recent publications demonstrate the effectiveness of active learning for NLP tasks such as Named Entity Recognition (NER) [37] and sentence classification [44]. Siddhant and Lipton [38] and Shen et.…”

Section: Scientific Corporamentioning

confidence: 99%

“…al. [37] compare several sampling strategies on NLP tasks and show that Maximum Normalized Log-Probability (MNLP) based on uncertainty sampling performs well in NER.…”

Section: Scientific Corporamentioning

confidence: 99%

Domain-Independent Extraction of Scientific Concepts from Research Articles

Brack

D’Souza

Hoppe

et al. 2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

We examine the novel task of domain-independent scientific concept extraction from abstracts of scholarly articles and present two contributions. First, we suggest a set of generic scientific concepts that have been identified in a systematic annotation process. This set of concepts is utilised to annotate a corpus of scientific abstracts from 10 domains of Science, Technology and Medicine at the phrasal level in a joint effort with domain experts. The resulting dataset is used in a set of benchmark experiments to (a) provide baseline performance for this task, (b) examine the transferability of concepts between domains. Second, we present two deep learning systems as baselines. In particular, we propose active learning to deal with different domains in our task. The experimental results show that (1) a substantial agreement is achievable by non-experts after consultation with domain experts, (2) the baseline system achieves a fairly high F1 score, (3) active learning enables us to nearly halve the amount of required training data.

show abstract

“…OntoNotes 5.0. On OntoNotes 5.0, we compare the proposed GRN with NER models that also reported performance on it, including (Chiu and Nichols 2016;Shen et al Model Mean(±std) F 1 Mean P/R (Chiu and Nichols 2016) 86.28 ± 0.26 86.04 / 86.53 (Shen et al 2017) 86.63 ± 0.49 (Durrett and Klein 2014) 84.04 85.22 / 82.89 (Passos, Kumar, and McCallum 2014) 82.30 (? ) 83.45 CNN-BiLSTM-Att-CRF 87.25±0.17 ID-CNN (Strubell et al 2017) 86.84 ± 0.19 GRN 87.67 ± 0.17 87.79 / 87.56 Table 3, GRN can obtain the state-of-the-art NER performance on OntoNotes 5.0, which further demonstrates its effectiveness.…”

Section: Performance Comparisonmentioning

confidence: 99%

GRN: Gated Relation Network to Enhance Convolutional Neural Network for Named Entity Recognition

Chen

Lin

Ding

et al. 2019

AAAI

View full text Add to dashboard Cite

The dominant approaches for named entity recognition (NER) mostly adopt complex recurrent neural networks (RNN), e.g., long-short-term-memory (LSTM). However, RNNs are limited by their recurrent nature in terms of computational efficiency. In contrast, convolutional neural networks (CNN) can fully exploit the GPU parallelism with their feedforward architectures. However, little attention has been paid to performing NER with CNNs, mainly owing to their difficulties in capturing the long-term context information in a sequence. In this paper, we propose a simple but effective CNN-based network for NER, i.e., gated relation network (GRN), which is more capable than common CNNs in capturing long-term context. Specifically, in GRN we firstly employ CNNs to explore the local context features of each word. Then we model the relations between words and use them as gates to fuse local context features into global ones for predicting labels. Without using recurrent layers that process a sentence in a sequential manner, our GRN allows computations to be performed in parallel across the entire sentence. Experiments on two benchmark NER datasets (i.e., CoNLL-2003 and Ontonotes 5.0) show that, our proposed GRN can achieve state-of-the-art performance with or without external knowledge. It also enjoys lower time costs to train and test. We have made the code publicly available at https://github.com

show abstract

Deep Active Learning for Named Entity Recognition

Cited by 308 publications

References 28 publications

Deep Reference Mining From Scholarly Literature in the Arts and Humanities

Deep Reference Mining From Scholarly Literature in the Arts and Humanities

Domain-Independent Extraction of Scientific Concepts from Research Articles

GRN: Gated Relation Network to Enhance Convolutional Neural Network for Named Entity Recognition

Contact Info

Product

Resources

About