Artificial intelligence has assumed a pivotal role in women’s healthcare. This research comprises three significant objectives: to investigate the contextual factors on students at a private university in Thailand who exhibit susceptibility to premenstrual symptoms and the potential for depressive episodes, to establish a risk classification model for premenstrual symptoms and the associated risk of depression among the student population at the aforementioned private university, and to examine the efficacy of the risk classification model concerning premenstrual symptoms and the correlated risk of depression within the same group of students. The collected data encompassed 375 students from Bangkokthonburi University in Thailand. The research instruments comprised two distinct types of medical health assessments: the DASS-21 and PSST-A assessments. The instruments employed to construct the model included random forest, gradient boosting, bagging, extra trees classification, voting, and stacking methodologies. The study findings indicate that the students at Bangkokthonburi University exhibit significantly positive mental and physical health. Meanwhile, the predictive model for the risk of menstrual depression is strong, so this research is worth promoting and developing into further applications.