Heat Map based Feature Ranker: In Depth Comparison with Popular Methods

Abstract : The new era of technology allows us to gather more data than ever before, complex data emerge and a lot of noise can be found among high dimensional datasets. In order to discard useless features and help build more generalized models, feature selection seeks a reduced subset of features that improve the performance of the learning algorithm. The evaluation of features and their interactions are an expensive process, hence the need for heuristics. In this work, we present HeatMap Based Feature Ranker, an algorithm to estimate feature importance purely based on its interaction with other variables. A compression mechanism reduces evaluation space up to 66% without compromising efficacy. Our experiments show that our proposal is very competitive against popular algorithms, producing stable results across different types of data. We also show how noise reduction through feature selection aids data visualization using emergent self-organizing maps.
Type de document :
Article dans une revue
Intelligent Data Analysis, IOS Press, In press, 22 (5), pp.1009-1037. 〈10.3233/IDA-173481〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01848544
Contributeur : Christian Raymond <>
Soumis le : mardi 24 juillet 2018 - 17:51:52
Dernière modification le : jeudi 15 novembre 2018 - 11:59:01
Document(s) archivé(s) le : jeudi 25 octobre 2018 - 16:34:33

Fichier

IDA218.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Carlos Huertas, Reyes Juárez-Ramírez, Christian Raymond. Heat Map based Feature Ranker: In Depth Comparison with Popular Methods. Intelligent Data Analysis, IOS Press, In press, 22 (5), pp.1009-1037. 〈10.3233/IDA-173481〉. 〈hal-01848544〉

Partager

Métriques

Consultations de la notice

115

Téléchargements de fichiers

93