Divorce Astrologer: Machine Learning based Divorce Prediction of Married Couples

The continuously growing number of divorce cases all across the world have become a serious matter of concern. Psychologists, counsellors and researchers have found a similar trend between the attitude of couples that lead to their divorce. Hence, there is a need of a specific algorithm-based prediction mechanism that takes all such factors into consideration and try to predict whether a couple will get divorced in the future. In this work, the authors present a rigorous research by comparing the standard existing machine learning algorithms with the ensemble algorithms and in turn comparing it to the existing related work in this field. The research talks about the implementation of Gottman’s method of divorce prediction using classification algorithms namely Support Vector Machine, Naïve Bayes, Decision Tree, Random Forest and Ensemble algorithms including Normal Gradient Boosting and ADA Boosting. The authors make sure that the best model of each algorithm is implemented in order to achieve the best accuracy in prediction. This system can be of great benefit to the psychologists and family counsellors to get a clear idea about the relationship status.