Student Depression Detection Dataset

Description

This dataset contains information about lifestyle, demographic, academic, and socio-economic characteristics of urban university students, along with a binary indicator reflecting whether the respondent reports being depressed with life. The dataset was designed to support research on mental health prediction, student lifestyle analysis, and machine learning–based depression detection. The dataset consists of 3,350 observations (rows) and 12 attributes (columns). Each row represents an individual respondent, while each column represents a demographic, academic, social, health, or financial factor potentially associated with mental well-being. The primary target variable is “Depressed with life”, which indicates whether a respondent experiences depression in life. The remaining attributes describe factors that may influence or correlate with depression among urban students. This dataset can be used for: Machine learning classification tasks Mental health prediction research Social and behavioral data analysis Feature importance analysis Student lifestyle and wellbeing studies Dataset Structure Feature List: Gender – Encoded gender of the respondent represented as a binary value. Age – Age of the respondent measured in years. Family Background – Indicates the socio-economic or family environment of the respondent. Relationship Status – Represents whether the respondent is currently in a romantic relationship. Marital Status – Describes the marital condition of the respondent (e.g., single or married). Living Style – Describes the living arrangement of the respondent, such as living with family, in a hostel, or in shared accommodation. Current Semester – Indicates the academic semester the respondent is currently enrolled in. CGPA – Cumulative Grade Point Average representing the respondent’s academic performance. Good Friend Circle – Indicates whether the respondent reports having a supportive or positive friend circle. Health Condition – Self-reported physical health status of the respondent. Per Month Cost – Estimated monthly personal expenditure of the respondent. Depressed with life – Target variable indicating whether the respondent reports experiencing depression in life (binary classification label). Dataset Characteristics: Total Samples: 3,350 Total Features: 12 Data Type: Mixed numerical and categorical (encoded numerically) Target Variable: Depressed with life Problem Type: Binary classification The dataset includes a combination of: Demographic attributes (age, gender, family background) Academic indicators (semester, CGPA) Social factors (friend circle, relationship status) Lifestyle attributes (living style, monthly expenditure) Health condition indicators

Institutions

• International University of Business Agriculture and Technology

  Dhaka Division, Dhaka

Categories

Licence