---
license: gpl-2.0
---
## Data Description

This is training data for machine learning models that can be used with [Cavil](https://github.com/openSUSE/cavil),
the openSUSE legal review and SBOM system.

Cavil uses a pattern matching system to identify potential legal text in source code. This process is based around identifying
hot zones of legal keywords (snippets) and produces around 80% false positives. Historically these false positives had to be
sorted out by humans. A few years ago we've started using machine learning to automate much of this process. And today we
train the model on these 150.000 samples, to reach about 96% accuracy.

There are currently two example implementations using this dataset:

1. https://github.com/kraih/Character-level-cnn-pytorch/
2. https://github.com/kraih/llm-lawyer

## Intended Use

This dataset is intended to be used to train machine learning models to identify legal text in Open Source code. It was curated by the humans of the SUSE legal review team.

## License

Licensed under [GPL-2.0-or-later](https://github.com/openSUSE/cavil/blob/master/COPYING).