Automatic Spelling Correction for Resource-Scarce Languages using Deep Learning

Etoori, Pravallika; Chinnakotla, Manoj Kumar; Mamidi, Radhika

doi:10.18653/v1/p18-3021

Cited by 44 publications

(26 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, there are some attempts to develop better spelling correction algorithms based on neural nets (Etoori et al, 2018). Similar to our baselines ScRNN (Sakaguchi et al, 2017) and MUDE , Li et al (2018) proposed a nested RNN to hierarchically encode characters to word representations, then correct each word using a nested GRU .…”

Section: Related Work and Backgroundmentioning

confidence: 99%

Context-aware Stand-alone Neural Spelling Correction

Li¹,

Liu

Huang

2020

Findings of the Association for Computational Linguistics: EMNLP 2020

View full text Add to dashboard Cite

Existing natural language processing systems are vulnerable to noisy inputs resulting from misspellings. On the contrary, humans can easily infer the corresponding correct words from their misspellings and surrounding context. Inspired by this, we address the standalone spelling correction problem, which only corrects the spelling of each token without additional token insertion or deletion, by utilizing both spelling information and global context representations. We present a simple yet powerful solution that jointly detects and corrects misspellings as a sequence labeling task by fine-turning a pre-trained language model. Our solution outperform the previous state-ofthe-art result by 12.8% absolute F 0.5 score.

show abstract

Section: Related Work and Backgroundmentioning

confidence: 99%

Context-aware Stand-alone Neural Spelling Correction

Li¹,

Liu

Huang

2020

Findings of the Association for Computational Linguistics: EMNLP 2020

View full text Add to dashboard Cite

show abstract

“…Considering technologies used for editing support, there have been many studies for various purposes, such as spelling error correction (Farra et al, 2014;Hasan et al, 2015;Etoori et al, 2018), grammatical error correction (Dahlmeier and Ng, 2012;Susanto et al, 2014;Choshen and Abend, 2018), fact checking (Baly et al, 2018;Thorne and Vlachos, 2018;Lee et al, 2018), fluency evaluation (Vadlapudi and Katragadda, 2010;Heilman et al, 2014;Kann et al, 2018), and so on. However, when we consider their studies on our task, they are only used after editing (writing a draft).…”

Section: Related Workmentioning

confidence: 99%

A Case Study on Neural Headline Generation for Editing Support

Murao¹,

Kobayashi²,

Kobayashi³

et al. 2019

Proceedings of the 2019 Conference of the North

View full text Add to dashboard Cite

There have been many studies on neural headline generation models trained with a lot of (article, headline) pairs. However, there are few situations for putting such models into practical use in the real world since news articles typically already have corresponding headlines. In this paper, we describe a practical use case of neural headline generation in a news aggregator, where dozens of professional editors constantly select important news articles and manually create their headlines, which are much shorter than the original headlines. Specifically, we show how to deploy our model to an editing support tool and report the results of comparing the behavior of the editors before and after the release.

show abstract

“…Recently, Sooraj et al (2018) employed a characterbased LSTM language model to detect and correct spelling errors for Malayalam. In the same line of research, Etoori et al (2018) propose an attention model with a bidirectional characterbased LSTM encoder-decoder trained end-to-end for Hindi and Telugu spelling correction using synthetic datasets.…”

Section: Related Workmentioning

confidence: 99%

Normalising Non-standardised Orthography in Algerian Code-switched User-generated Data

Adouane

Bernardy

Dobnik

2019

Proceedings of the 5th Workshop on Noisy User-Generated Text (W-Nut 2019)

View full text Add to dashboard Cite

We work with Algerian, an under-resourced non-standardised Arabic variety, for which we compile a new parallel corpus consisting of user-generated textual data matched with normalised and corrected human annotations following data-driven and our linguistically motivated standard. We use an end-toend deep neural model designed to deal with context-dependent spelling correction and normalisation. Results indicate that a model with two CNN sub-network encoders and an LSTM decoder performs the best, and that word context matters. Additionally, preprocessing data token-by-token with an editdistance based aligner significantly improves the performance. We get promising results for the spelling correction and normalisation, as a pre-processing step for downstream tasks, on detecting binary Semantic Textual Similarity.

show abstract

Automatic Spelling Correction for Resource-Scarce Languages using Deep Learning

Cited by 44 publications

References 12 publications

Context-aware Stand-alone Neural Spelling Correction

Context-aware Stand-alone Neural Spelling Correction

A Case Study on Neural Headline Generation for Editing Support

Normalising Non-standardised Orthography in Algerian Code-switched User-generated Data

Contact Info

Product

Resources

About