Topological Feature Selection

Antonio Briola, Tomaso Aste
2/19/2023
cs.LGcs.AI

No official code found. Best implementation: FinancialComputingUCL/Topological_Feature_Selection (8 stars, last commit 2023-02-21).

Abstract

In this paper, we introduce a novel unsupervised, graph-based filter feature selection technique which exploits the power of topologically constrained network representations. We model dependency structures among features using a family of chordal graphs (the Triangulated Maximally Filtered Graph), and we maximise the likelihood of features' relevance by studying their relative position inside the network. Such an approach presents three aspects that are particularly satisfactory compared to its alternatives: (i) it is highly tunable and easily adaptable to the nature of input data; (ii) it is fully explainable, maintaining, at the same time, a remarkable level of simplicity; (iii) it is computationally cheaper compared to its alternatives. We test our algorithm on 16 benchmark datasets from different applicative domains showing that it outperforms or matches the current state-of-the-art under heterogeneous evaluation conditions.

View on arXivView PDF

Code Implementations(1)

FinancialComputingUCL/Topological_Feature_SelectionCommunity100%

In this repository we present a novel unsupervised, graph-based filter feature selection technique which exploits the power of topologically constrained network representations.

82PythonFeb 17, 20233 years ago
complex-networksfeature-selectioninformation-filteringnetwork-sciencenetworks

Similar Papers

Enhancing Graph Representation Learning with Localized Topological Features

80% match1 repo182025

Graph-based Extreme Feature Selection for Multi-class Classification Tasks

80% match2023

Chordless cycle filtrations for dimensionality detection in complex networks via topological data analysis

79% match2025

Is Graph Convolution Always Beneficial For Every Feature?

78% match2024

Mitigating Label Noise on Graph via Topological Sample Selection

78% match1 repo42026

Mitigating Label Noise on Graph via Topological Sample Selection

78% match2 repos42024

Cite this paper

@article{briola2023topological,
  title  = {Topological Feature Selection},
  author = {Antonio Briola and Tomaso Aste},
  year   = {2023},
  eprint = {2302.09543},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2302.09543v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2302.09543.svg)](https://codeofpaper.com/paper/2302.09543)

Discussion