Unbalanced Timeliness of Financial Reporting Data Classification Using Random Forest with SMOTE

This study aims to apply the Random Forest method with SMOTE to address unbalanced data on company classifications based on the timeliness of financial reports. The data used are the financial statements of manufacturing companies in the Food and Beverage sector on the IDX from 2014 to 2022. The independent variables used are ROA, CR, DAR, and Size. The results showed that the performance of the Random Forest method after being combined with SMOTE increased compared to before SMOTE. Random Forest's best performance is derived from 60% training and 40% testing. Based on MDA and MDG values, it was found that ROA has the highest level of importance, followed by Size and CR variables. In comparison, DAR is the variable with the lowest level of importance. It means that DAR has a low impact on the timeliness of financial reports.