Part-of-Speech Tagging from 97% to 100%: Is It Time for Some Linguistics?

By Christopher D. Manning. 2011

449 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cultural Studies

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Cultural Studies

Citations in the corpus

End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF (2016, methodology): “In order to compare with previous work, we adopt the standard splits — section 0-18 as training data, section 1921 as development data and section 22-24 as test data (Manning, 2011; Søgaard, 2011).”
Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics (2020, neutral): “We hypothesize that hard-to-learn examples—those with low conﬁdence — might be mislabeled, as has also been suggested in prior work (Manning, 2011; Toneva et al., 2018).”
Natural Language Processing Advancements By Deep Learning: A Survey (2020, methodology): “Traditional methods are still performing very well for this task [12].”
Robust Multilingual Part-of-Speech Tagging via Adversarial Training (2017, methodology): “The per token accuracy metric is easy since taggers can easily assign correct POS tags to highly unambiguous tokens, such as punctuation (Manning, 2011).”
Learning Character-level Representations for Part-of-Speech Tagging (2014, methodology): “The result of our system for the WSJ corpus is similar to the ones of (Manning, 2011) and (Collobert et al., 2011), and very competitive with the result of (Søgaard, 2011).”
SenticNet 3: A Common and Common-Sense Knowledge Base for Cognition-Driven Sentiment Analysis (2014, methodology): “Labels, in particular, are useful in case we deal with real-time adaptive applications (in which, for example, the style of an interface or the expression of an avatar has to quickly change according to user's input).”
Empower Sequence Labeling with Task-Aware Neural Language Model (2017, methodology): “We also list some top reported performance on each dataset from (Collobert et al. 2011; Luo et al. 2015; Chiu and Nichols 2016; Yang, Salakhutdinov, and Cohen 2017; Peters et al. 2017; Manning 2011; Søgaard and Goldberg 2016; Sun 2014) for reference.”
A Challenge Set and Methods for Noun-Verb Ambiguity (2018, methodology): “Measurement Manning (2011) performed an error analysis for WSJ and discovered that 19% of the errors fall under "Difﬁcult linguistics" which need non-local context modeling to be able to solve them.”
Multilingual Part-of-Speech Tagging with Bidirectional Long Short-Term Memory Models and Auxiliary Loss (2016, methodology): “We also report accuracies on WSJ (45 POS) using the standard splits (Collins, 2002; Manning, 2011).”
XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization (2020, methodology): “For POS tagging, we adopt 97% as a canonical estimate of human performance based on Manning (2011).”
We Need to Talk about Standard Splits (2019, methodology): “• Stanford tagger (Manning, 2011): a loglinear bidirectional cyclic dependency network, features from Toutanova et al.”
A Universal Part-of-Speech Tagset (2011, background): “…supervised POS tagging accuracies for English (measured on the Wall Street Journal portion of the Pen-nTreebank (Marcus et al., 1993)) have converged to around 97.3% (Toutanova et al., 2003; Shen et al., 2007), the attention has shifted to unsupervised approaches (Christodoulopoulos et al., 2010).”