English

Unsupervised Feature Ranking via Attribute Networks

Machine Learning 2021-11-29 v1

Abstract

The need for learning from unlabeled data is increasing in contemporary machine learning. Methods for unsupervised feature ranking, which identify the most important features in such data are thus gaining attention, and so are their applications in studying high throughput biological experiments or user bases for recommender systems. We propose FRANe (Feature Ranking via Attribute Networks), an unsupervised algorithm capable of finding key features in given unlabeled data set. FRANe is based on ideas from network reconstruction and network analysis. FRANe performs better than state-of-the-art competitors, as we empirically demonstrate on a large collection of benchmarks. Moreover, we provide the time complexity analysis of FRANe further demonstrating its scalability. Finally, FRANe offers as the result the interpretable relational structures used to derive the feature importances.

Keywords

Cite

@article{arxiv.2111.13273,
  title  = {Unsupervised Feature Ranking via Attribute Networks},
  author = {Urh Primožič and Blaž Škrlj and Sašo Džeroski and Matej Petković},
  journal= {arXiv preprint arXiv:2111.13273},
  year   = {2021}
}

Comments

for online material and python package see https://github.com/FRANe-team/FRANe

R2 v1 2026-06-24T07:52:33.293Z