Mind Matters: USDI-30 Depression Dataset of Bangladeshi University Students for Machine Learning Analysis

The Mind Matters: USDI-30 Depression Dataset of Bangladeshi University Students for Machine Learning Analysis contains data from 823 university students, collected from two prominent institutions in Bangladesh: Daffodil International University (DIU) and the University of Dhaka. The dataset is based on the University Student Depression Inventory (USDI-30) scale, which measures 30 key psychological and emotional indicators of depression, such as mood, motivation, self-worth, and social engagement.

Each participant's data was collected with individual informed consent, ensuring that all respondents were aware of the purpose of the study, the confidentiality of their data, and their right to withdraw at any time. Demographic information was also collected with consent, including age, gender, academic year, and field of study, providing richer context for analysis.

To further enhance the dataset’s validity, all responses were reviewed and validated by domain experts, including licensed psychologists, ensuring that the data accurately reflects the mental health status of the respondents.

Key features:

1. 823 validated responses from DIU and University of Dhaka students.
2. 30 depression indicators based on the USDI-30 scale, covering comprehensive mental health dimensions.
3. Demographic data including age, gender, academic year, and field of study, collected with informed consent.
4. Domain expert validation by psychologists, ensuring high data quality and relevance.
5. The dataset is cleaned and pre-processed, making it ready for machine learning and predictive analysis.

Depression Labelling Steps: The answers of USDI-30 questions converted into numerical score such as Never = 0, Rarely (less than one day) = 1, Occasionally (1-2 days) = 2, Frequently (3-4 days) = 3, and Most of the time (5-7 days) = 4.

The depression levels based on the USDI Scale:
1. 0-30 = No Depression
2. 30-60 = Mild Depression
3. 60-90 = Moderate Depression
4. 90-120 = Severe Depression

This dataset serves as a valuable resource for mental health researchers, data scientists, and AI practitioners. It provides a solid foundation for developing predictive models, understanding depression trends, and facilitating early intervention strategies for university students at risk of depression.

《心灵关照》：孟加拉国大学生抑郁症数据集（USDI-30）旨在为机器学习分析提供数据，该数据集收录了来自孟加拉国两家著名学府——达芙妮国际大学（DIU）和达卡大学——的823名大学生的数据。本数据集基于大学生抑郁症量表（USDI-30），该量表用于衡量抑郁症的30项关键心理和情绪指标，如情绪、动机、自尊和社会参与度等。（University Student Depression Inventory (USDI-30)）。

每一位参与者的数据均是在知情同意的前提下收集的，确保所有受访者均了解研究目的、数据保密性以及他们在任何时间都有权撤回同意。同时，在同意的基础上，还收集了人口统计学信息，包括年龄、性别、学年以及研究领域，为分析提供了更丰富的背景信息。

为了进一步提升数据集的有效性，所有回答均经过领域专家的审查和验证，包括执业心理学家，确保数据准确反映了受访者的心理健康状况。

关键特性：
1. 来自DIU和达卡大学学生的823份经过验证的回复。
2. 基于USDI-30量表，包含30项抑郁症指标，全面覆盖心理健康维度。
3. 收集了年龄、性别、学年以及研究领域等人口统计学数据，均是在知情同意的基础上进行的。
4. 由心理学家进行的领域专家验证，确保数据质量高且相关性强。
5. 数据集经过清理和预处理，使其适用于机器学习和预测分析。

抑郁症标记步骤：将USDI-30问题的答案转换为数值分数，例如“从未”= 0，“很少（少于一天）”= 1，“偶尔（1-2天）”= 2，“经常（3-4天）”= 3，以及“大多数时间（5-7天）”= 4。

基于USDI量表的抑郁症水平：
1. 0-30 = 无抑郁症
2. 30-60 = 轻度抑郁症
3. 60-90 = 中度抑郁症
4. 90-120 = 严重抑郁症

本数据集为心理健康研究人员、数据科学家和AI从业者提供了一个宝贵的资源，它为开发预测模型、理解抑郁症趋势以及为有抑郁症风险的大学生提供早期干预策略奠定了坚实基础。