Analyzing STEM Women’s Employability: A Machine Learning Framework with Clustering and Multi-Target Classification

The gender gap in Science, Technology, Engineering, and Mathematics (STEM) fields highlights significant research opportunities, particularly in examining the employability of female graduates. This study introduces a novel machine learning framework integrating Clustering and Multi-target Classification to analyze employment waiting time and job linearity among women STEM alumni. Using K-Means Clustering and Multi-Target Logistic Regression, the framework achieved a classification accuracy of 77% and a silhouette score of 0.61, demonstrating its effectiveness in predictive analysis. Beyond these results, the framework offers a robust methodology for integrating Clustering with Classification, enabling a nuanced understanding of the employability challenges faced by women in STEM. This approach identifies key patterns in employment data, paving the way for targeted interventions and actionable insights. Furthermore, the findings aim to inform data-driven policymaking and future research to improve employability outcomes. This work contributes to addressing systemic challenges and fostering gender diversity in STEM careers while enhancing opportunities for women