<![CDATA[Monitoring the state of labor data in Indonesia involves standardized classification to ensure uniformity. The coding process relies on the knowledge of personnel, which often leads to issues such as potential differences in understanding and interpretation among individuals, resulting in inconsistencies in the standardized classification coding outcomes. This study aims to explore the potential of Machine Learning in classifying the Indonesia Business Field Classification (KBLI) and the Indonesian Standard Classification of Occupations (KBJI). Models were developed and evaluated to classify KBLI and KBJI based on open-ended questions about the job, the output produced, and the field of work from respondents' answers collected through the National Labor Force Survey (Sakernas). The results show that although the performance of the IndoBERT method is slightly superior with accuracy is 0,76 for KBLI and 0,65 for KBJI. This advantage is not significant given the higher computational load and longer training time compared to machine learning.]]>
Copyrights © 2024
