PySS3 is a Python package for text classification, offering an interpretable machine learning model (SS3) with built-in visualization tools for Explainable AI. It's designed for researchers and practitioners who need to understand and trust their classification models, particularly in sensitive applications.

How It Works

The core SS3 model is a novel, white-box classifier that naturally explains its rationale. It achieves interpretability by directly exposing the features and their weights contributing to a classification decision. PySS3 enhances this with a Live_Test component for interactive, visual debugging and an Evaluation class for automated hyperparameter tuning and performance visualization via interactive 3D plots.

Quick Start & Requirements

• Install via pip: pip install pyss3
• Requires Python.
• Live demos and full documentation are available online.

Highlighted Details

• Achieved top results in CLEF's eRisk lab for early depression detection.
• Supports multi-label classification.
• extract_insight() method provides text fragments driving classification.
• Evaluation.plot() generates portable HTML files for sharing interactive performance visualizations.

Maintenance & Community

The project acknowledges several contributors following the all-contributors specification. Contact is available via email for further information.

Licensing & Compatibility

The README does not explicitly state a license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The project's license is not clearly stated, which may impact commercial adoption. Specific hardware or software requirements beyond Python are not detailed.