Maternal Depression Dataset: A Machine Learning Perspective on the Mental Health Analysis of Mothers of School-Going Children in Dhaka using the Zung SDS Scale
Description
This dataset comprises socio-demographic attributes and psychological depression indicators, collected using the Zung Self-Rating Depression Scale (SDS). The study specifically focuses on mothers of school-going children in Dhaka, Bangladesh, aiming to evaluate their mental health conditions. Data Collection Methodology: Initially, the dataset was collected from 12 schools in Dhaka, selecting mothers of children enrolled in public and private institutions. Participants were chosen randomly, ensuring representation from diverse educational backgrounds. The inclusion criteria required mothers of children from classes V to VIII who had attended the same school for at least six months. Additionally, data was collected from the Daffodil Institute of Social Sciences (DISS) to enhance the dataset’s diversity. The dataset contains 594 records classified into four depression severity levels: 1. Not Depressed 2. Mild Depressed 3. Moderate Depressed 4. Severe Depressed Features: The dataset integrates socio-demographic information, including age range, marital status, education level, and occupation, along with mental health indicators such as: - Emotional Well-being: Downhearted feelings, crying spells, hopelessness - Physiological Symptoms: Sleep disturbances, appetite changes, weight loss, increased heart rate - Cognitive Patterns: Trouble concentrating, decision-making difficulty, restlessness - Suicidal Ideation: Feelings of being "better off dead" Applications in Machine Learning & Health Informatics: - Predictive modeling for maternal depression assessment - Explainable AI (XAI) for interpretable mental health diagnosis - mHealth applications for early detection and intervention - Statistical and behavioral analysis of psychological distress