Predicting Depression and Anxiety in Women Using LDA-Based CNN
Abstract
Depression affects women of all ages in India. The stress of balancing many roles can lead to depression in Indian women, which goes untreated due to social stigma. Among the most prevalent mental health conditions that affect women in their reproductive years are premenstrual dysphoric syndrome, postpartum depression, and PMS. Primary care physicians should emphasize early diagnosis of intimate relationships and domestic abuse and mandate routine tests for these issues. Since antidepressants constitute the cornerstone of treatment, they ought to be freely available at all primary care levels. When individuals take their medication as directed for a sufficient period and keep in regular contact with mental health professionals, the best possible results are obtained. The best results are obtained when cognitive therapy is used in conjunction with other non-pharmacological methods. In this work, deep learning architectures using Linear Discriminant Analysis (LDA) were utilized. Possible contributions to the studies include convolutional neural networks (CNNs) and transformer-based pre-trained language models for classification. When the six functional status groups are utilized rather than just one set of depressed symptoms, the results will be more dependable and consistent.