A comparison of methods for modeling soundscape dimensions based on different datasets

  • Soundscape studies vary considerably in study design, statistical methods, and model fit metrics used. Due to this confounding of data and methods, it is difficult to assess the suitability of statistical modelling techniques used in the literature. Therefore, five different methods and two performance metrics were applied to three existing soundscape datasets to model soundscape Pleasantness and Eventfulness based on seven acoustic and three sociodemographic predictors. Datasets differed in soundscape type (urban outdoor vs indoor), experimental setting (field- vs lab-based), size, and study design (site- vs person-centered). The fixed-effects and mixed-effects methods ranged from linear to nonlinear regression based on advanced machine learning approaches. Results showed that models performed better for Eventfulness than for Pleasantness in most cases, while performance as measured by the out-of-sample R2 was dependent on the total variance of the target, especially in both fieldSoundscape studies vary considerably in study design, statistical methods, and model fit metrics used. Due to this confounding of data and methods, it is difficult to assess the suitability of statistical modelling techniques used in the literature. Therefore, five different methods and two performance metrics were applied to three existing soundscape datasets to model soundscape Pleasantness and Eventfulness based on seven acoustic and three sociodemographic predictors. Datasets differed in soundscape type (urban outdoor vs indoor), experimental setting (field- vs lab-based), size, and study design (site- vs person-centered). The fixed-effects and mixed-effects methods ranged from linear to nonlinear regression based on advanced machine learning approaches. Results showed that models performed better for Eventfulness than for Pleasantness in most cases, while performance as measured by the out-of-sample R2 was dependent on the total variance of the target, especially in both field studies with imbalanced targets and groups. Nonlinear methods consistently outperformed linear regression, with random forest and extreme gradient boosting performing particularly well, while the performance levels of all nonlinear methods remained comparable. Mixed-effects models provided a more generalized, albeit slightly smaller prediction performance when tested on unknown groups. Finally, this study motivates the use of cross-validation with special splitting for analyzing small imbalanced datasets.show moreshow less

Download full text files

Export metadata

Metadaten
Author:Siegbert VersümerORCiD, Patrick Blättermann, Fabian RosenthalORCiD, Stefan WeinzierlORCiD
Qualitätssicherung:peer reviewed
open access:Hybrid - Subskriptionszeitschrift
Fachbereich/Einrichtung:Hochschule Düsseldorf / Fachbereich - Medien
Document Type:Article
Year of Completion:2025
Language of Publication:English
Publisher:AIP Publishing
Parent Title (English):The Journal of the Acoustical Society of America
Volume:157
Issue:1
First Page:234
Last Page:255
URN:urn:nbn:de:hbz:due62-opus-48849
DOI:https://doi.org/10.1121/10.0034849
ISSN:1520-8524
Tag:Acoustic ecology; Acoustical properties; Artificial neural networks; Audiometry; Finite-element analysis; Machine learning; Psychological acoustics; Regression analysis; Statistical analysis; Urban development
GND Keyword:Akustische Eigenschaft; Audiometrie; Psychoakustik; Neuronales Netz; Maschinelles Lernen; Stadtentwicklung; Finite-Elemente-Methode; Regressionsanalyse; Statistische Analyse
Corresponding Author:Siegbert Versümer
Information on the Research Data:The data that support the findings of this study are available within the article and its supplementary material. The analysis framework developed as part of this study is freely available as the Python package flexcv at https://doi.org/10.5281/zenodo.10160846 and https://github.com/radlfabs/flexcv. The data we have re-evaluated have been made publicly available by their authors or can be requested.
Dewey Decimal Classification:0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 000 Informatik, Informationswissenschaft, allgemeine Werke
Licence (German):Creative Commons - CC BY - Namensnennung 4.0 International
Release Date:2025/01/22
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.