Explaining automated gender classification of human gait
- Publication type:
- Journal article
- Metadata:
-
- Autoren
- F Horst
- D Slijepcevic
- M Zeppelzauer
- AM Raberger
- S Lapuschkin
- W Samek
- WI Schöllhorn
- C Breiteneder
- B Horsak
- DOI
- 10.1016/j.gaitpost.2020.07.114
- ISSN
- 0966-6362
- Zeitschrift
- Gait & Posture
- Sprache
- en
- Paginierung
- 159 - 160
- Datum der Veröffentlichung
- 2020
- Status
- Published
- Herausgeber
- Elsevier BV
- Herausgeber URL
- http://dx.doi.org/10.1016/j.gaitpost.2020.07.114
- Datum der Datenerfassung
- 2020
- Titel
- Explaining automated gender classification of human gait
- Ausgabe der Zeitschrift
- 81
Data source: Crossref
- Other metadata sources:
-
- Author's licence
- InCopyright
- Autoren
- Fabian Horst
- Djordje Slijepcevic
- Matthias Zeppelzauer
- Anna-Maria Raberger
- Sebastian Lapuschkin
- Wojciech Samek
- Wolfgang I Schöllhorn
- Christian Breiteneder
- Brian Horsak
- Hosting institution
- Universitätsbibliothek Mainz
- Resource version
- Published version
- DOI
- 10.1016/j.gaitpost.2020.07.114
- File(s) embargoed
- false
- Open access
- true
- ISSN
- 0966-6362
- Ausgabe der Veröffentlichung
- Supplement 1
- Zeitschrift
- Gait & Posture
- Sprache
- eng
- Notes
- 3 pages, 1 figure
- Open access status
- Open Access
- Paginierung
- 159 - 160
- Zugang
- Deleted
- Titel
- Explaining automated gender classification of human gait
- Ausgabe der Zeitschrift
- 81
Files
explaining_automated_gender_c-20230628092418905.pdf
Data source: OPENSCIENCE.UB
- Abstract
- State-of-the-art machine learning (ML) models are highly effective in classifying gait analysis data, however, they lack in providing explanations for their predictions. This "black-box" characteristic makes it impossible to understand on which input patterns, ML models base their predictions. The present study investigates whether Explainable Artificial Intelligence methods, i.e., Layer-wise Relevance Propagation (LRP), can be useful to enhance the explainability of ML predictions in gait classification. The research question was: Which input patterns are most relevant for an automated gender classification model and do they correspond to characteristics identified in the literature? We utilized a subset of the GAITREC dataset containing five bilateral ground reaction force (GRF) recordings per person during barefoot walking of 62 healthy participants: 34 females and 28 males. Each input signal (right and left side) was min-max normalized before concatenation and fed into a multi-layer Convolutional Neural Network (CNN). The classification accuracy was obtained over a stratified ten-fold cross-validation. To identify gender-specific patterns, the input relevance scores were derived using LRP. The mean classification accuracy of the CNN with 83.3% showed a clear superiority over the zero-rule baseline of 54.8%.
- Autoren
- Fabian Horst
- Djordje Slijepcevic
- Matthias Zeppelzauer
- Anna-Maria Raberger
- Sebastian Lapuschkin
- Wojciech Samek
- Wolfgang I Schöllhorn
- Christian Breiteneder
- Brian Horsak
- Autoren-URL
- http://arxiv.org/abs/2211.17015v1
- Zeitschrift
- Gait & Posture 81 (Supplement 1) (2020) 159-160
- Schlüsselwörter
- cs.LG
- cs.LG
- Notes
- 3 pages, 1 figure
- Datum der Veröffentlichung
- 2022
- Herausgeber URL
- http://dx.doi.org/10.1016/j.gaitpost.2020.07.114
- Datum der Datenerfassung
- 2022
- Datum, an dem der Datensatz öffentlich gemacht wurde
- 2022
- Titel
- Explaining automated gender classification of human gait
Files
2211.17015v1.pdf
Data source: arXiv
- Beziehungen:
- Property of