A Systematic Literature Review on Bengali Grammar Correction System

Abstract Around 270 million people worldwide speak Bangla, making it the sixth most widely spoken language. Despite its widespread use, the study of the Bengali language was behind schedule. There has been a significant lag in the application of recently introduced technologies, such as transformer-based models, to the Bengali language. Pretrained models such as GPT3 and Llama2 do not perform effectively when it comes to the Bengali language due to its distinctive linguistic characteristics. Our initial step in improving the Bengali language inside the realm of state-of-the-art technology involves analysing prior automated grammar correction systems and comprehending the challenges specific to the Bengali language. This article offers a thorough examination of prior research on automated grammar correction systems. In order to gain a deeper understanding of the prior cutting-edge techniques, strategies, and challenges in this particular domain, this review seeks to consolidate and analyze the pertinent literature in a discerning manner. Using a pre-established search methodology, we successfully identified relevant research in reputable databases and scholarly publications. The review provides a concise overview of the findings from the selected research and highlights the diverse range of techniques employed, such as rule-based tactics, statistical models, machine learning algorithms, and natural language processing methods. The analysis also includes an evaluation of crucial indicators of system performance, including accuracy, recall, and F1-score. Through this extensive literature review, we discovered recurring challenges in developing grammatical correction algorithms that possess both accuracy and contextual sensitivity, which aligns with our subsequent objective. Furthermore, based on these findings, this study considers future research objectives and prospects for improving automatic grammar correction technologies. To summarise, this systematic literature review provides a comprehensive examination of the existing automatic grammar correction systems, highlighting their accomplishments, the challenges that remain, and the potential for further investigation and advancement in this area. This will provide us with valuable insights to align Bengali language research with the present setting. Additionally, other researchers, developers, and educators can greatly benefit from the information presented here as they strive to create and implement a more efficient Bengali grammatical correction system or beyond.