Geodesic
HMC-ReliefF: Feature Ranking for Hierarchical Multi-label Classification
Computer Science and Information Systems, Tome 15 (2018) no. 1.

Voir la notice de l'article provenant de la source Computer Science and Information Systems website

In machine learning, the growing complexity of the available data poses an increased challenge for its analysis. The rising complexity is both in terms of the data becoming more high-dimensional as well as the data having a more intricate structure. This emphasizes the need for developing machine learning algorithms that are able to tackle both the high-dimensionality and the complex structure of the data. Our work in this paper focuses on the development and analysis of the HMCReliefF algorithm, which is a feature relevance (ranking) algorithm for the task of Hierarchical Multi-label Classification (HMC). The basis of the algorithm is the RReliefF algorithm for regression that is adapted for hierarchical multi-label target variables. We perform an extensive experimental investigation of the HMC-ReliefF algorithm on several datasets from the domains of image annotation and functional genomics. We analyse the algorithm’s performance in terms of accuracy in a filterlike setting and also in terms of ranking stability for various parameter values. The results show that the HMC-ReliefF can successfully detect relevant features from the data that can be further used for constructing accurate predictive models. Additionally, the stability analysis helps to determine the preferred parameter values for obtaining not just accurate, but also a stable algorithm output.
Keywords: feature selection, feature ranking, structured data, hierarchical multilabel classification, ReliefF 
@article{CSIS_2018_15_1_a8,
     author = {Ivica Slavkov and Jana Karcheska and Dragi Kocev and Sa\v{s}o D\v{z}eroski},
     title = {HMC-ReliefF: {Feature} {Ranking} for {Hierarchical} {Multi-label} {Classification}},
     journal = {Computer Science and Information Systems},
     publisher = {mathdoc},
     volume = {15},
     number = {1},
     year = {2018},
     url = {http://geodesic.mathdoc.fr/item/CSIS_2018_15_1_a8/}
}
TY  - JOUR
AU  - Ivica Slavkov
AU  - Jana Karcheska
AU  - Dragi Kocev
AU  - Sašo Džeroski
TI  - HMC-ReliefF: Feature Ranking for Hierarchical Multi-label Classification
JO  - Computer Science and Information Systems
PY  - 2018
VL  - 15
IS  - 1
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/CSIS_2018_15_1_a8/
ID  - CSIS_2018_15_1_a8
ER  - 
%0 Journal Article
%A Ivica Slavkov
%A Jana Karcheska
%A Dragi Kocev
%A Sašo Džeroski
%T HMC-ReliefF: Feature Ranking for Hierarchical Multi-label Classification
%J Computer Science and Information Systems
%D 2018
%V 15
%N 1
%I mathdoc
%U http://geodesic.mathdoc.fr/item/CSIS_2018_15_1_a8/
%F CSIS_2018_15_1_a8
Ivica Slavkov; Jana Karcheska; Dragi Kocev; Sašo Džeroski. HMC-ReliefF: Feature Ranking for Hierarchical Multi-label Classification. Computer Science and Information Systems, Tome 15 (2018) no. 1. http://geodesic.mathdoc.fr/item/CSIS_2018_15_1_a8/