This study developed a typo correction system for the Indonesian language by integrating the Levenshtein Distance algorithm with empirical methods. The system is designed to improve the accuracy of typo detection and correction in Indonesian texts, which feature complex morphological structures such as prefixes, suffixes, and compound words. The findings show that the system achieved a precision rate of 92% and an F1-score of 90.5%, indicating high reliability in providing relevant correction suggestions. Additionally, the system demonstrated efficiency in processing time, with an average of 0.8 seconds for short texts and 5.3 seconds for longer texts. The use of empirical methods enables the system to handle complex language variations, resulting in more contextually appropriate correction suggestions. User feedback indicated high satisfaction with the interface and the relevance of the suggestions provided. Overall, this research makes a significant contribution to the development of more adaptive and efficient typo correction systems for the Indonesian language and opens up opportunities for further development in the context of other similar languages.
