Spelling Error Patterns in Brazilian Portuguese

Gimenes, Priscila A.; Roman, Norton Trevisan; Carvalho, Ariadne Maria Brito Rizzoni

doi:10.1162/coli_a_00216

Cited by 8 publications

(7 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some writing systems (such as Arabic, Vietnamese, or Slovak) use different character variants that change the meaning of the word. The authors in [10] confirmed that the omission of diacritics is a common type of spelling error in Brazilian Portuguese. Texts in Modern Standard Arabic are typically written without diacritical markings [11].…”

Section: Spelling Errorsmentioning

confidence: 69%

Survey of Automatic Spelling Correction

2020

View full text Add to dashboard Cite

Automatic spelling correction has been receiving sustained research attention. Although each article contains a brief introduction to the topic, there is a lack of work that would summarize the theoretical framework and provide an overview of the approaches developed so far. Our survey selected papers about spelling correction indexed in Scopus and Web of Science from 1991 to 2019. The first group uses a set of rules designed in advance. The second group uses an additional model of context. The third group of automatic spelling correction systems in the survey can adapt its model to the given problem. The summary tables show the application area, language, string metrics, and context model for each system. The survey describes selected approaches in a common theoretical framework based on Shannon’s noisy channel. A separate section describes evaluation methods and benchmarks.

show abstract

Section: Spelling Errorsmentioning

confidence: 69%

Survey of Automatic Spelling Correction

2020

View full text Add to dashboard Cite

show abstract

“…However, these rules are exclusively applicable to English, so they could not be generalized to other languages. In recent years, studies have also been conducted on error patterns for other languages, namely Portuguese (Gimenes, Roman & Carvalho, 2015), Hungarian (Siklósi, Novák & Prószéky, 2016), Japanese (Baba & Suzuki, 2012), Danish (Paggio, 2000), and Punjabi (Lehal & Bhagat, 2007).…”

Section: Error Typologiesmentioning

confidence: 99%

Linguistic errors in the biomedical domain: Towards an error typology for Spanish

Hernández¹,

Almela²,

Valencia‐García³

2021

View full text Add to dashboard Cite

Linguistic errors in the biomedical domain: Towards an error typology for Spanish. The objective of this work is the analysis of errors contained in a corpus of medical reports in natural language and the design of a typology of errors, as there was no systematic review on verification and correction of errors in clinical documentation in Spanish. In the development of automatic detection and correction systems, it is of great interest to delve into the nature of the linguistic errors that occur in clinical reports, in order to detect and treat them properly. The results show that omission errors are the most frequent ones in the analyzed sample, and that word length certainly influences error frequency. The typification of error patterns provided is enabling the development of a module based on linguistic knowledge, which is currently in progress. This will help to improve the performance of error detection and correction systems for the biomedical domain.

show abstract

“…Building misspelling detector and serve it before corrector is essential, however detection compared with correction is much less discussed, and the importance of applying detection in real production associated with the most widely adopted noisy-channel approach is even yet disclosed. Some works discuss misspelling detection, correction and pattern analysis from language perspective, such as Arabic [3], Portuguese [10], Persian [26], however the methods are either around noisy-channel paradigm or only language model or mainly the pattern analysis. We also see related discussions for domain-specific applications such as for clinical text [16][25] [26], and indeed search [17][25] [20] as well, however they either focus on patterns or improving a specific area under the noisy-channel architecture such as efficient dictionary or matching technique.…”

Section: Related Workmentioning

confidence: 99%

Maps Search Misspelling Detection Leveraging Domain-Augmented Contextual Representations

Li¹

2021

Preprint

View full text Add to dashboard Cite

Building an independent misspelling detector and serve it before correction can bring multiple benefits to speller and other search components, which is particularly true for the most commonly deployed noisy-channel based speller systems. With rapid development of deep learning and substantial advancement in contextual representation learning such as BERTology 1 , building a decent misspelling detector without having to rely on hand-crafted features associated with noisy-channel architecture becomes morethan-ever accessible. However BERTolgy models are trained with natural language corpus but Maps Search is highly domain specific, would BERTology continue its success. In this paper we design 4 stages of models for misspeling detection ranging from the most basic LSTM to single-domain augmented fine-tuned BERT. We found for Maps Search in our case, other advanced BERTology family model such as RoBERTa does not necessarily outperform BERT, and a classic cross-domain fine-tuned full BERT even underperforms a smaller single-domain fine-tuned BERT. We share more findings through comprehensive modeling experiments and analysis, we also briefly cover the data generation algorithm breakthrough. CCS CONCEPTS• Information systems → Query intent; • Computing methodologies → Neural networks.

show abstract

Spelling Error Patterns in Brazilian Portuguese

Cited by 8 publications

References 9 publications

Survey of Automatic Spelling Correction

Survey of Automatic Spelling Correction

Linguistic errors in the biomedical domain: Towards an error typology for Spanish

Maps Search Misspelling Detection Leveraging Domain-Augmented Contextual Representations

Contact Info

Product

Resources

About