<![CDATA[Spelling errors are a common problem in text processing, one of which is in Indonesian. The increasing use of non-standard language, especially in digital text communication, is the background for this research. Spelling errors in sensitive religious content can even cause misunderstandings. This article examines the development of a model for spelling correction with a deep learning-based approach using Sequence-to-Sequence with a GRU-based encoder-decoder architecture and attention mechanism. A dataset containing standard and non-standard text pairs is used to test the model. The experimental results show that the proposed model produces 76.82% accuracy, but is able to recognize and correct spelling errors. However, this research is expected to contribute in the future, so that it can improve improvements in the Indonesian spelling correction system.]]>
Copyrights © 2024
