Application of AI to formal methods

This repository contains a data set of 457 scientific publications which apply artificial intelligence onto the area of formal methods. The data were aggregated within scope of a systematic mapping study which can be found here.

For the five year span of 2019-2023 consisting of 189 publications, we conducted a mapping study and sorted the entries according to their utilized AI technique and their respective application domain within the formal methods research area.

We provide the following files:

• database.bib: The full list of found publications, in Bibtex format. Processed entries have a groups field attached. See also groups.csv.
• overview/ai-techniques_2019-2023.md: List of the processed publications from 2019-2023, grouped by their utilized AI technique.
• overview/fm-techniques_2019-2023.md: List of the processed publications from 2019-2023, grouped by their utilized FM technique.
• overview/all_by_year.md: List of all entries divided by their respective publication years.
• overview/data-sets_2019-2023.md: List of all found, published data sets and RL environments to train models.

More information on how we assembled the data can be found in our systematic mapping study.

Cite this repository

If you are finding use in our aggregated database of primary studies within your own research, please consider using the following reference.

• tbd.

@tbd{}

Build notes

We build the repository files with Python 3.10+. You can optionally choose to prepare a virtual environment. Dependencies are stored in requirements.txt as usual.

# Setup virtual environment (optional but recommended)
python3.10 -m venv env
. env/bin/activate
# Install dependencies
pip install -r requirements.txt

If you want to update the underlying search results, run the preprocessing first. The preprocessed version is checked in for easier use by external researchers. The respective search_results.bib and not_in_time_frame.bib files are available in the dev-only repository of the article.

python -m dev search_results.bib not_in_time_frame.bib > database.bib

To update the overview files, edit make_overviews.py to your liking and run it.

python make_overviews.py