Infinite feature selection: a graph-based feature filtering approach

Roffo, G. , Melzi, S., Castellani, U., Vinciarelli, A. and Cristani, M. (2020) Infinite feature selection: a graph-based feature filtering approach. IEEE Transactions on Pattern Analysis and Machine Intelligence, (doi: 10.1109/TPAMI.2020.3002843) (Early Online Publication)

[img]
Preview
Text
218830.pdf - Accepted Version

2MB

Abstract

We propose a filtering feature selection framework that considers a subset of features as a path in a graph, where a node is a feature and an edge indicates pairwise (customizable) relations among features, dealing with relevance and redundancy principles. By two different interpretations (exploiting properties of power series of matrices and relying on Markov chains fundamentals) we can evaluate the values of paths (i.e., feature subsets) of arbitrary lengths, eventually go to infinite, from which we dub our framework Infinite Feature Selection (Inf-FS). Going to infinite allows to constrain the computational complexity of the selection process, and to rank the features in an elegant way, that is, considering the value of any path (subset) containing a particular feature. We also propose a simple unsupervised strategy to cut the ranking, so providing the subset of features to keep. In the experiments, we analyze diverse setups with heterogeneous features, for a total of 11 benchmarks, comparing against 18 widely-known yet effective comparative approaches. The results show that Inf-FS behaves better in almost any situation, that is, when the number of features to keep are fixed a priori, or when the decision of the subset cardinality is part of the process.

Item Type:Articles
Status:Early Online Publication
Refereed:Yes
Glasgow Author(s) Enlighten ID:Roffo, Dr Giorgio and Vinciarelli, Professor Alessandro
Authors: Roffo, G., Melzi, S., Castellani, U., Vinciarelli, A., and Cristani, M.
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
College/School:College of Science and Engineering > School of Computing Science
Research Group:Social Signal Processing
Journal Name:IEEE Transactions on Pattern Analysis and Machine Intelligence
Publisher:IEEE
ISSN:0162-8828
ISSN (Online):1939-3539
Published Online:16 June 2020
Copyright Holders:Copyright © 2020 IEEE
First Published:First published in IEEE Transactions on Pattern Analysis and Machine Intelligence 2020
Publisher Policy:Reproduced in accordance with the publisher copyright policy
Related URLs:

University Staff: Request a correction | Enlighten Editors: Update this record

Project CodeAward NoProject NamePrincipal InvestigatorFunder's NameFunder RefLead Dept
172889A Robot Training Buddy for adults with ASDAlessandro VinciarelliEngineering and Physical Sciences Research Council (EPSRC)EP/N035305/1Computing Science