Paper Title
Detection of Postpartum Psychological Disorders By Using Swarm Intelligence Techniques: A Comparative Analysis of GWO And ALO

Abstract
This manuscript presents the detection of postpartum psychological disorders using two different swarm intelligence techniques viz. Ant Lion Optimization (ALO) and Grey Wolf Optimization (GWO). The principal objective of swarm based optimization technique is to refine feature selection process by enhancing the percentage of predictive accuracy of machine learning models. Such type of approach is helpful for medical and healthcare applications because critical clinical decisions are based on accurate predictions which are ultimately responsible for better clinical intervention and diagnostic support. The classification of postpartum psychological disorders is based on the symptomatic manifestations observed in the women during and after childbirth. The central focus of this study is the early identification of medical conditions related to mental health care, with the goal of enhancing psychological well-being. In this study, “ALO” and “GWO” algorithms are employed for feature selection process. The simulation environment indicates that the selected feature functions serve as input variables, where the predictor or target column represents the output—shows whether the female population is affected by postpartum psychological disorder or not. Hence, by utilizing swarm intelligence algorithms for feature selection, the accuracy percentage of different machine learning models can be significantly improved. The simulation results of the proposed methodology indicate that RF optimized with the Grey Wolf Optimization Algorithm (GWOA) achieved an accuracy of 95%, while XGBoost(GWO) reached 96%. In comparison, RF optimized with the Ant Lion Optimization Algorithm (ALOA) achieved an accuracy of 93%, and XGBoost(ALO) attained 92%. Keywords - Postpartum Psychological Disorders, Swarm Intelligence, Machine Learning, Ant Lion Optimization (ALO), Grey Wolf Optimization (GWO), Performance Metrics, Random Forest (RF) Classifier, XGBoost Classifier.