Data-driven MSME Success Prediction Using Decision Tree-Based Machine Learning Techniques

Abstract

MSMEs play an important role in Indonesia's digital economy, but not all businesses are able to survive and thrive sustainably. The low predictive ability of MSME success is a major challenge in formulating effective policies and interventions. This research aims to build a prediction model for the success of MSMEs by utilizing machine learning algorithms as a strategic decision-making tool. The approach used is an experimental method by comparing the performance of three popular algorithms: Decision Tree, Random Forest, and Support Vector Machine (SVM). The dataset used comes from a combination of survey and open source data, which includes variables of MSME characteristics. The data was analyzed through preprocessing, model training, and evaluation using accuracy, precision, recall, and F1-score metrics. The results showed that the Random Forest algorithm performed best with an optimal classification balance. This finding indicates that the machine learning approach is effective in identifying MSME success patterns based on historical data. The main contribution of this research is the development of an artificial intelligence-based decision support system that can be adapted for the local context to support the sustainable growth of digital MSMEs.