TY - JOUR

T1 - Learning complementary representations via attention-based ensemble learning for cough-based COVID-19 recognition

AU - Ren, Zhao

AU - Chang, Yi

AU - Nejdl, Wolfgang

AU - Schuller, Björn W.

N1 - Funding Information: This work was partially supported by the BMBF project LeibnizKILabor with grant No. 01DD20003, and the Horizon H2020 Marie Skõõlodowska-Curie Actions Initial Training Network European Training Network (MSCA-ITN-ETN) project under grant agreement No. 766287 (TAPAS). The authors would thank the friendly discussions from their colleagues Lukas Stappen and Vincent Karas.

PY - 2022/7/25

Y1 - 2022/7/25

N2 - Coughs sounds have shown promising as-potential marker for distinguishing COVID individuals from non-COVID ones. In this paper, we propose an attention-based ensemble learning approach to learn complementary representations from cough samples. Unlike most traditional schemes such as mere maxing or averaging, the proposed approach fairly considers the contribution of the representation generated by each single model. The attention mechanism is further investigated at the feature level and the decision level. Evaluated on the Track-1 test set of the DiCOVA challenge 2021, the experimental results demonstrate that the proposed feature-level attention-based ensemble learning achieves the best performance (Area Under Curve, AUC: 77.96%), resulting in an 8.05% improvement over the challenge baseline.

AB - Coughs sounds have shown promising as-potential marker for distinguishing COVID individuals from non-COVID ones. In this paper, we propose an attention-based ensemble learning approach to learn complementary representations from cough samples. Unlike most traditional schemes such as mere maxing or averaging, the proposed approach fairly considers the contribution of the representation generated by each single model. The attention mechanism is further investigated at the feature level and the decision level. Evaluated on the Track-1 test set of the DiCOVA challenge 2021, the experimental results demonstrate that the proposed feature-level attention-based ensemble learning achieves the best performance (Area Under Curve, AUC: 77.96%), resulting in an 8.05% improvement over the challenge baseline.

KW - Attention mechanism

KW - Complementary representation

KW - Cough sound

KW - COVID-19

KW - Ensemble learning

UR - http://www.scopus.com/inward/record.url?scp=85135094432&partnerID=8YFLogxK

U2 - 10.1051/aacus/2022029

DO - 10.1051/aacus/2022029

M3 - Article

AN - SCOPUS:85135094432

VL - 6

JO - Acta Acustica

JF - Acta Acustica

IS - 29

M1 - 29

ER -