PyAerial Documentation

Welcome to PyAerial’s documentation!

PyAerial is a Python implementation of the Aerial scalable neurosymbolic association rule miner for tabular data.

It aims to learn a small set of high-quality rules as opposed to exhaustive rule miners such as Apriori, FP-Growth, ECLAT etc. This is done by learning a neural (network) represantation of the given table, which captures only the most prominent patterns, and then the rules are extracted from the neural representation.

Learn more about the architecture, training, and rule extraction in the paper: Neurosymbolic Association Rule Mining from Tabular Data (NeSy 2025), or in the How Aerial Works page.

Performance

PyAerial significantly outperforms traditional ARM methods in scalability, also on CPU, while maintaining high-quality results:

PyAerial performance comparison

Execution time comparison across datasets of varying sizes. PyAerial scales linearly while traditional methods (e.g., Mlxtend, SPMF) exhibit exponential growth.

Key advantages:

  • ⚡ 100-1000x faster on large datasets compared to Apriori/FP-Growth Python implementations

  • 📈 Linear scaling in training, polynomial scaling in rule extraction

  • 🎯 No rule explosion - extracts concise, high-quality rules with full data coverage

  • 💾 Memory efficient - neural representation avoids storing exponential candidate sets

  • 🖥️ Fast on CPU - GPU is optional and only needed for very large datasets

For comprehensive benchmarks and comparisons with Mlxtend (e.g., FPGrowth, Apriori etc.), and other ARM tools, see our benchmarking paper: PyAerial: Scalable association rule mining from tabular data (SoftwareX, 2025).

Contents:

Indices and tables