PHQ-9 Student Depression Dataset

The PHQ-9 Student Depression Dataset contains responses from 500 students to the PHQ-9 questionnaire, a well-established tool for diagnosing depression. This dataset is designed to support the development of machine learning models aimed at automated depression detection by analyzing text responses to common depression-related questions.

The PHQ-9 questionnaire includes 9 questions that assess symptoms of depression over the past two weeks, covering areas like mood, energy levels, sleep, appetite, and thoughts of self-harm. The responses are scored on a scale from 0 (Not at all) to 3 (Nearly every day), with the total score ranging from 0 to 27. Based on this score, the depression severity is classified into one of the following categories:
Minimal (0-4)
Mild (5-9)
Moderate (10-14)
Moderately Severe (15-19)
Severe (20-27)

This dataset is primarily designed for building models that can assist in automated depression detection. Some potential use cases include:
Sentiment Analysis: Analyzing emotional tones in text responses to assess depression.
Text Classification: Classifying responses into different depression severity levels.
Predictive Modeling: Predicting depression severity based on textual responses.
Feature Engineering: Extracting linguistic features (e.g., sentiment, keywords) to predict depression.
The dataset is diverse, with synthetic responses across different levels of depression, providing a versatile foundation for machine learning applications. While the dataset does not contain personally identifiable information (PII), real-world applications should follow ethical guidelines regarding privacy, consent, and mental health resources.
When working with real data or applying this dataset in clinical research, it is essential to adhere to ethical standards, including:

Data Privacy: Anonymizing personal information.
Informed Consent: Ensuring participants give consent before data collection.
Support Resources: Providing support for individuals who may exhibit serious mental health concerns.

Applications:
Clinical Research: This dataset is valuable for studying depression detection using natural language processing and machine learning techniques.
AI in Healthcare: It can be used in the development of tools for automated mental health assessment.
Education: Training students or professionals in recognizing depression symptoms and analyzing responses.

PHQ-9学生抑郁症数据集收录了500名学生针对患者健康问卷-9（PHQ-9）的作答数据，该问卷是临床广泛使用的成熟抑郁症诊断工具。本数据集旨在助力基于机器学习的自动化抑郁症检测模型研发，模型可通过分析受试者针对抑郁症相关常规问题的文本作答实现检测。

PHQ-9问卷共包含9个问题，用于评估受试者过去两周内的抑郁症状，覆盖情绪、精力水平、睡眠、食欲以及自残意念等多个维度。作答得分采用0至3分的李克特量表：0分代表「完全没有」，3分代表「几乎每日出现」；总得分区间为0至27分，据此可将抑郁严重程度划分为以下类别：
- 极轻微抑郁（0-4分）
- 轻度抑郁（5-9分）
- 中度抑郁（10-14分）
- 中重度抑郁（15-19分）
- 重度抑郁（20-27分）

本数据集主要用于研发可辅助自动化抑郁症检测的模型，其潜在应用场景包括：
1. 情感分析：分析作答文本的情感基调以评估抑郁状态
2. 文本分类：将作答文本划分为不同的抑郁严重程度类别
3. 预测建模：基于作答文本预测抑郁严重程度
4. 特征工程：提取语言特征（如情感、关键词）以实现抑郁预测

本数据集覆盖不同抑郁严重程度的合成作答样本，具备良好的多样性，可为机器学习相关应用提供通用的研发基础。尽管本数据集不包含个人可识别信息（PII），但在实际应用中仍需遵循隐私、知情同意及心理健康资源相关的伦理准则。

若使用真实数据或在临床研究中应用本数据集，则必须遵守以下伦理规范：
- 数据隐私：对个人信息进行匿名化处理
- 知情同意：确保受试者在数据收集前已签署知情同意书
- 支持资源：为可能存在严重心理健康问题的受试者提供援助支持

应用场景：
1. 临床研究：本数据集可用于基于自然语言处理与机器学习技术的抑郁症检测研究
2. 医疗人工智能：可用于研发自动化心理健康评估工具
3. 教育领域：可用于培训学生或专业人员识别抑郁症状并分析作答文本

搜集汇总

背景与挑战

背景概述

PHQ-9学生抑郁症数据集包含682名17-26岁学生的PHQ-9问卷回答，是第5版增强数据集，新增了心理社会因素变量，数据收集过程符合伦理标准，适用于抑郁症研究和机器学习分析。