This repository contains the experimental source code for the paper entitled "Discovering Association Rules in High-Dimensional Small Tabular Data", accepted at the 1st International Workshop on Advanced Neuro-Symbolic Applications as part of the European Conference on Artificial Intelligence 2025.
- Datasets
- Baseline implementations
- Installing dependencies
- Running scalability experiments on high-dimensional small tabular datasets
- Running fine-tuning experiments
- References
The datasets [1, 2, 3, 4] used in the experiments can be found under the data/discrete folder. Please see the
references for their original papers.
The datasets are pre-processed according the procedure described in [5], and the pre-processing is (partially) done by [6]. The pre-processing consists of the trimmed mean of m-values normalization, log transformation (i.e., $log( x+1)$), and the expression values were made to have zero mean and unit standard deviation.
We further pre-process the data to enable ARM on gene expression datasets and applied z-score binning with one standard deviation as the cutoff to discretize values into high, low, and medium gene expression levels.
The source code for discretization is given
under src/data_preparation/discretization.py
- Aerial+ [7] is implemented using PyAerial [12] library.
- Fine-tuned versions of Aerial+, as proposed in our paper, can be found
in
src/fine_tuned_aerial/model_double_loss_pretrained.py( Aerial+DL) and undersrc/experiments/fine-tuning_via_pretrained_weight_initialization.py( Aerial+WI). - FP-Growth [8] is implemented using MLxtend.
- ECLAT [9] is implemented using pyECLAT.
- ARM-AE [10] source code is taken from its original repository.
First, create a python virtual environment:
python3 -m venv venv
Activate the virtual environment:
source venv/bin/activate
Install the requirements:
pip3 install -r requirements.txt
Install TabPFN [11] extensions package:
pip install "tabpfn-extensions[all] @ git+https://github.com/PriorLabs/tabpfn-extensions.git"
The scalability experiments in Section 3 of our paper can be run as follows:
python3 -m src.experiments.scalability_on_high_dimensional_data
The results are both printed to stdout, as well as saved into a .csv file named "scalability_experiments_ {timestamp}.csv":
The proposed fine-tuning methods in Section 4 can be run as follows.
Aerial+WI implementation described in our paper can be found under src/experiments/fine-tuning_via_pretrained_weight_initialization.py. The code can be run via:
python3 -m src.experiments.fine-tuning_via_pretrained_weight_initialization
The results are printed to stdout, including the final average:
Aerial+DL implementation is in src/experiments/fine-tuning_via_double_loss.py. The code can be run via:
python3 -m src.experiments.fine-tuning_via_double_loss
The results are printed to stdout, including the final average:
[1] W. Yang, J. Soares, P. Greninger, E. J. Edelman, H. Lightfoot, S. Forbes, N. Bindal, D. Beare, J. A. Smith, I. R. Thompson, et al. Genomics of drug sensitivity in cancer (gdsc): a resource for therapeutic biomarker discovery in cancer cells. Nucleic acids research, 41(D1):D955–D961, 2012.
[2] F. Iorio, T. A. Knijnenburg, D. J. Vis, G. R. Bignell, M. P. Menden, M. Schubert, N. Aben, E. Gonçalves, S. Barthorpe, H. Lightfoot, et al. A landscape of pharmacogenomic interactions in cancer. Cell, 166(3): 740–754, 2016
[3] H. Gao, J. M. Korn, S. Ferretti, J. E. Monahan, Y. Wang, M. Singh, C. Zhang, C. Schnell, G. Yang, Y. Zhang, et al. High-throughput screening using patient-derived tumor xenografts to predict clinical trial drug response. Nature medicine, 21(11):1318–1325, 2015.435
[4] M. J. Garnett, E. J. Edelman, S. J. Heidorn, C. D. Greenman, A. Dastur, K. W. Lau, P. Greninger, I. R. Thompson, X. Luo, J. Soares, et al. Systematic identification of genomic markers of drug sensitivity in cancer cells. Nature, 483( 7391):570–575, 2012.
[5] S. M. Mourragui, M. Loog, D. J. Vis, K. Moore, A. G. Manjon, M. A. van de Wiel, M. J. Reinders, and L. F. Wessels. Predicting patient response with models trained on cell lines and patient-derived xenografts by nonlinear transfer learning. Proceedings of the National Academy of Sciences, 118(49):e2106682118, 2021.
[6] C. Ruiz, H. Ren, K. Huang, and J. Leskovec. High dimensional, tabular deep learning with an auxiliary knowledge graph. Advances in Neural Information Processing Systems, 36:26348–26371, 2023.
[7] E. Karabulut, P. Groth, and V. Degeler. Neurosymbolic association rule mining from tabular data. In Proceedings of the 19th Conference on Neurosymbolic Learning and Reasoning (NeSy 2025), Accepted/In Press. Preprint available at https://arxiv.org/abs/2504.19354.
[8] Han, J. Pei, and Y. Yin. Mining frequent patterns without candidate generation. ACM sigmod record, 29(2):1–12, 2000.
[9] M. J. Zaki, S. Parthasarathy, M. Ogihara, W. Li, et al. New algorithms for fast discovery of association rules. In KDD, volume 97, pages 283–286, 1997.
[10] T. Berteloot, R. Khoury, and A. Durand. Association rules mining with auto-encoders. In International Conference on Intelligent Data Engineering and Automated Learning, pages 51–62. Springer, 2024.
[11] N. Hollmann, S. Müller, L. Purucker, A. Krishnakumar, M. Körfer, S. B. Hoo, R. T. Schirrmeister, and F. Hutter. Accurate predictions on small data with a tabular foundation model. Nature, 637(8045):319–326, 2025.
[12] Erkan Karabulut, Paul Groth, and Victoria Degeler. Pyaerial: Scalable association rule mining from tabular data. SoftwareX, 31:102341, 2025b. ISSN 2352-7110. doi:https://doi.org/10.1016/j.softx.2025.102341.