Mathematical modeling of intelligent system for predicting effectiveness of premenstrual syndrome

Abstract

Majority of reproductive-aged women experience some form of physical discomfort or emotional unease in the weeks leading up to the onset of menstruation. The symptoms are often not severe, but they can cause significant discomfort and disrupt day to day activities of the person experiencing them. It is estimated that between 5 and 8 percent of women experience severe premenstrual syndrome (PMS); the majority of these women may also fall under the category of premenstrual dysphoric disorder (PMDD). The most bothersome symptoms are those associated with the mood and behaviour, such as impatience, tension, sad mood, tearfulness, and mood swings. However, physical problems, such as breast soreness, indigestion and bloating, can also be problematic. Using the Gradian Boost regressor (GBR) method of machine learning, the researchers in this study made a prediction regarding the effects of premenstrual syndrome (PMS). Kelly Wallance classifies premenstrual syndrome as PMS-A, PMS-C, PMS-D, and PMS-H, in addition to other symptoms. Researchers circulated the Kelly Wallance questionnaire on Google Form, which was then used to collect the data for the dataset. The accuracy of the model was measured at 99.99% for PMS-A, 99.93% for PMS-C, 99.87% for PMS-D, 99.92% for PMS-H, and 99.97% for other symptoms.