Bangla Spelling Error Detection and Correction Using N-Gram Model

Bagchi, Promita; Arafin, Mursalin; Akther, Aysha; Alam, Kazi Masudul

doi:10.1007/978-3-031-34619-4_37

Cited by 1 publication

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach aimed to create a comprehensive dataset that authentically reflects the varied types of changes necessary for Bengali language learners and users. The entire annotation process has been carried out by three native Bengali language teachers (L1), who were appointed through Surge AI 1 and possess expertise in Bengali language. To ensure the quality and authenticity of the corrections, we engaged four other proficient Bengali language experts native to West Bengal and Bangladesh.…”

Section: Datasetmentioning

confidence: 99%

“…Despite the growing interest in GEC and the availability of GEC datasets in high-resource languages such as English [11,6,20], Chinese [32], German [4], Russian [24], Spanish [12], etc., there is a noticeable scarcity of real-world GEC datasets specifically tailored for low-resource languages such as Bengali. Although there is existing GEC research for Bengali [23,27,19,18,1], no work has been undertaken in the areas of feedback or explanation generation within this context. A notable effort by [10] in feedback comment generation (FCG) introduces a typology for learning feedback, covering abstract types (e.g., tone and idiom) and grammatical pattern types (e.g., comparative and causative) in the English language.…”

Section: Related Workmentioning

confidence: 99%

“…[3], current works [23,27,19,18,1] in Bengali GEC encounter a significant challenge in providing comprehensive explanations for errors in natural language alongside their corrections. The provision of error explanations is pivotal for effective language learning and teaching, as emphasized by [14].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

How Ready Are Generative Pre-trained Large Language Models for Explaining Bengali Grammatical Errors?

Maity,

Deroy,

Sarkar

2024

Preprint

View full text Add to dashboard Cite

Grammatical error correction (GEC) tools, powered by advanced generative artificial intelligence (AI), competently correct linguistic inaccuracies in user input. However, they often fall short in providing essential natural language explanations, which are crucial for learning languages and gaining a deeper understanding of the grammatical rules. There is limited exploration of these tools in low-resource languages such as Bengali. In such languages, grammatical error explanation (GEE) systems should not only correct sentences but also provide explanations for errors. This comprehensive approach can help language learners in their quest for proficiency. Our work introduces a real-world, multi-domain dataset sourced from Bengali speakers of varying proficiency levels and linguistic complexities. This dataset serves as an evaluation benchmark for GEE systems, allowing them to use context information to generate meaningful explanations and high-quality corrections. Various generative pre-trained large language models (LLMs), including GPT-4 Turbo, GPT-3.5 Turbo, Text-davinci-003, Text-babbage-001, Text-curie-001, Text-ada-001, Llama-2-7b, Llama-2-13b, and Llama-2-70b, are assessed against human experts for performance comparison. Our research underscores the limitations in the automatic deployment of current state-of-the-art generative pre-trained LLMs for Bengali GEE. Advocating for human intervention, our findings propose incorporating manual checks to address grammatical errors and improve feedback quality. This approach presents a more suitable strategy to refine the GEC tools in Bengali, emphasizing the educational aspect of language learning.

show abstract

Section: Datasetmentioning

confidence: 99%