Hybrid of Ensemble Machine Learning and Nature-Inspired Algorithms for Divorce Prediction

Divorce is a global issue with profound emotional, psychological, and socio-economic consequences. In 2022, Addis Ababa witnessed 14,000 registered marriages but also recorded 1,623 divorces, while 2018 saw 1,923 divorces. Understanding the factors contributing to divorce is vital for prevention and support. Machine learning and AI play a critical role in predicting divorce, early marital distress detection, and personalized interventions. Their scalability aids in effective prevention strategies and targeted support. This research explores a Hybrid Approach AdaBoost, Gradient Boosting, Bagging, Stacking, XGBoost, and Random Forest with Jaya and Whale Optimization. The Hybrid Approach is chosen to synergize the strengths of ensemble learning and nature-inspired optimization algorithms. The goal is to enhance divorce prediction accuracy by leveraging ensemble models' robustness and optimization inspired by natural processes. To assess model performance, train-test splits, and k-fold Cross-Validation techniques are used, with metrics like accuracy, precision, recall, F1 score, and AUC-ROC(Area Under the Receiver Operating Characteristic Curve). AdaBoost stands out, achieving 97%, and 96% accuracy in Jaya andWOA hyperparameter optimizations, respectively. This research aligns with Sustainable Development Goals (SDGs) by promoting gender equality (SDG 5), identifying inequalities and offering targeted support (SDG 10), and fostering stable families and social cohesion (SDG 16). By leveraging AI for divorce prediction, this work contributes to a more sustainable and inclusive world, advancing gender equality, reducing inequalities, and peaceful societies.