PsyNLP

A Multi-Modal NLP System for Privacy-Preserving Mental Health Text Analysis

Overview

PsyNLP is a web-based application that analyzes text (articles, journals, notes) for emotional content and potential mental health indicators. It uses transformer-based models to detect emotions and assess mental health severity.

Features

• Emotion Detection: Identify emotions in text using BERT-based models
• Mental Health Analysis: Detect potential mental health concerns
• Severity Scoring: Calculate overall severity scores for analyzed text
• LLM Enhancement: Optional suicidal risk assessment using local LLMs
• HTML Reports: Generate detailed visual reports with charts
• Web Interface: User-friendly Streamlit web application
• File Support: Process .txt, .md, and .docx files

Deployment (Automated)

If you have Python 3.12 installed, you can use the automated deployment script:

python deploy.py

This script will handle dependencies installation and model downloading.

After deployment, run the app with:

uv run streamlit run app.py

Installation (for non MacOS system)

1. Clone the repository:

git clone https://github.com/HackingU0/PsyNLP.git
cd PsyNLP

2. Install dependencies:

pip install -r requirements.txt

3. Download required models (They should be downloaded with deploy script)

Installation (for macOS)

xcode-select --install
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
git clone https://github.com/HackingU0/PsyNLP.git
brew install python@3.12
cd PsyNLP
python deploy.py

Usage

1. Start the web application:

uv run streamlit run app.py

2. The application should open automatically in your browser. If not, navigate to the URL shown in the terminal (usually http://localhost:8501).

3. Upload a text file

4. View the analysis results and generated report

Project Structure

PsyNLP/
├── modules/              # Core analysis modules
│   ├── file_reader.py
│   ├── sentence_process.py
│   ├── predict_nlp.py
│   ├── predict_score.py
│   ├── llm_enhancements.py
│   └── visualizer.py
├── nlp_models/          # Pre-trained models
├── templates/           # HTML templates
├── uploads/             # Uploaded files
├── reports/             # Generated reports
└── app.py              # Main Streamlit application

Requirements

• Python 3.12 (Does not support 3.13)
• PyTorch
• Transformers
• Streamlit
• spaCy
• See requirements.txt for full list

License

See LICENSE file for details.

Disclaimer

This tool is for educational and research purposes only. It should not be used as a substitute for professional mental health assessment or treatment.