Discovering Association Rules in High-Dimensional Small Tabular Data

Erkan Karabulut, Daniel Daza, Paul Groth, Victoria Degeler · 2025

Read Discovering Association Rules in High-Dimensional Small Tabular Data by Erkan Karabulut, Daniel Daza, Paul Groth, Victoria Degeler (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-20113.txt.

Description

Erkan Karabulut and three co-authors wrote this 2025 arXiv paper on data mining. Association rule mining looks for simple if-then patterns between features in a dataset. Such rules support knowledge discovery and models that people can inspect. In wide datasets, though, the number of rules explodes and searches grow costly.

The authors study Aerial+, a neural-symbolic method built to tame that explosion. They show it scales far better than leading algorithmic and neural baselines on five real datasets. They also define a new hard setting: wide data with very few rows, like gene data with thousands of features and dozens of samples. Two fine-tuning tricks that use tabular foundation models then improve rule quality. The gains held across the five datasets they tested.

Catalog record

CREATORErkan Karabulut, Daniel Daza, Paul Groth, Victoria Degeler
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE36 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-20113

Browse the full library · About the Tower of Babel