Have a Cake and Eat it Too: Assessing Discrimination Performance of an Intelligibility Index Obtained from a Reduced Sample Size - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Have a Cake and Eat it Too: Assessing Discrimination Performance of an Intelligibility Index Obtained from a Reduced Sample Size

Anna K Marczyk
  • Fonction : Auteur
  • PersonId : 1041539
A Ghio
Muriel Lalain
Marie Rebourg
C. Fredouille

Résumé

This paper investigates random vs. phonetically motivated reduction of linguistic material used in an intelligibility task in speech disordered populations and the subsequent impact on the discrimination classifier quantified by the area under the receiver operating characteristics curve (AUC of ROC). The comparison of obtained accuracy indexes shows that when the sample size is reduced based on a phonetic criterium-here, related to phonotactic complexity-, the classifier has a higher ranking ability than when the linguistic material is arbitrarily reduced. Crucially, downsizing the linguistic sample to about 30% of the original dataset does not diminish the discriminatory performance of the classifier. This result is of significant interest to both clinicians and patients as it validates a tool that is both reliable and efficient.
Fichier principal
Vignette du fichier
2020.lrec-1.220.pdf (281.2 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...

Dates et versions

hal-02611678 , version 1 (18-05-2020)

Identifiants

  • HAL Id : hal-02611678 , version 1

Citer

Anna K Marczyk, A Ghio, Muriel Lalain, Marie Rebourg, C. Fredouille, et al.. Have a Cake and Eat it Too: Assessing Discrimination Performance of an Intelligibility Index Obtained from a Reduced Sample Size. 12th Conference on Language Resources and Evaluation (LREC 2020), ELRA, May 2020, Marseille, France. pp.1784-1788. ⟨hal-02611678⟩
103 Consultations
60 Téléchargements

Partager

Gmail Facebook X LinkedIn More