<![CDATA[The proportion of individuals categorized as Not in Education, Employment, or Training (NEET) is one of the key indicators of the success of creative digital economic development among the youth. This study investigates NEET status on the island on Java island in 2023 using a machine learning approach. Despite Java being the economic and infrastructural center of Indonesia, there exist significant disparities in NEET rates across its provinces. These disparities reflect unequal access to education and employment opportunities, thereby hindering the achievement of Sustainable Development Goal (SDG) 8. By employing the XGBoost algorithm, this study successfully developed a classification model with exceptional performance. The XGBoost model, optimized through SMOTENN resampling and hyperparameter tuning, achieved a validation accuracy of 98.69%, a training loss of 0.0320, a validation loss of 0.0491, and a ROC-AUC score of 0.9978. These results represent a substantial improvement over the baseline model, which attained an accuracy of approximately 80%. The findings reveal that the primary factors influencing NEET status include age, marital status, education level, work experience, household size, gender, disability status, and training experience. Furthermore, participation in training programs and residence in urban areas are associated with a lower risk of becoming NEET, as they enhance individual skill sets and facilitate greater access to educational and employment opportunities.]]>
