Detection and Enhancement of Bahasa Standardized Word in Online News Draft using Levenshtein Algorithm :U-Tapis

The research on standardized word error detection and correction is based on technological advances and artificial intelligence as well as the problem of errors that are common in writing journalistic articles due to time constraints and work demands. The research on standardized word error detection and correction is also based on previous research, namely U-Tapis research. The research is a development of U-Tapis research to create a text-based application machine learning to detect and correct word or sentence errors in Indonesian. The research uses the Levenshtein Distance algorithm as the main algorithm for processing input articles. This algorithm will match each character in a word and produce a ratio or similarity value. The research shows success with the accuracy of nonstandard word detection up to 99% and F1 score reaching 92.5%. The research proved to be effective in detecting nonstandard words in Bahasa news articles and was able to replace the word with a standard word.