AI for Healthcare: A Classification Model for Personalized Premenstrual Symptoms and Depressive Crisis Risk Tracking Using Data Analytics and Machine Learning

A significant obstacle to women's lives is menstruation, which, in some cases, leads to depression, anxiety, and stress. This research, therefore, aims to solve this problem with three objectives: to study the context of female adolescents who experience premenstrual syndrome (PMS) problems and depression crises, to develop a classification model for personalized premenstrual symptoms and depressive crisis risk tracking using data analytics and machine learning techniques, and to evaluate the performance of all developed models to select suitable models for future application. The research data included 282 volunteers who gave consent from the University of Phayao, with random sampling. The machine learning used in this research consists of six techniques: decision trees, k-nearest neighbors, logistic regression, naïve bayes, random forests, and support vector machines. The model performance is tested using 10-fold cross-validation and evaluated using five metrics: accuracy, precision, recall, f1-score, and time. The results of the construction and evaluation of all models are at a high level, which is encouraging. The researchers expect that this research can be further developed for future mobile applications.