<p>This study investigates the economic consequences of divorce for women through the application of machine learning techniques to large-scale longitudinal data. Despite significant progress in gender equality in recent decades, divorce continues to have asymmetric financial impacts, with women often experiencing greater economic vulnerability post-dissolution. Using a dataset of 4328 divorced women across diverse socioeconomic backgrounds, we applied supervised learning algorithms to identify predictive factors of financial instability and recovery trajectories. Our findings reveal that educational attainment, pre-divorce occupation, child custody arrangements, and alimony outcomes significantly predict post-divorce financial trajectories. Random forest models achieved 83.7% accuracy in predicting which women would experience severe income decline (&gt; 40%) within two years post-divorce. Additionally, cluster analysis identified four distinct financial adaptation patterns, each associated with different sociodemographic profiles. We conclude by recommending targeted policy interventions and financial education programs to mitigate economic vulnerability for at-risk divorced women.</p>