A Comparative Analysis of Clustering and Feature Extraction Methods for the Automated Construction of Bird Species Classification Datasets DOI Open Access

Virgínia A. dos Santos,

Diego T. Terasaka,

Luiz E. R. Martins

et al.

Published: Nov. 17, 2024

The identification of bird species enables the creation machine learning models that can be employed for non-invasive monitoring populations. In this study, we present an advancement in assisted automated a training set classification species, with specific focus on Pantanal. Typically, process is conducted manually, which highly time-consuming approach. phase, propose comprehensive comparative testing to ascertain optimal methodologies feature extraction and clustering. Five clustering methods four were subjected testing. results our experiments demonstrate method purpose work was hierarchical clustering, using BirdNET extraction. This combination provided superior performance classifying construction sets.

Language: Английский

A Comparative Analysis of Clustering and Feature Extraction Methods for the Automated Construction of Bird Species Classification Datasets DOI Open Access

Virgínia A. dos Santos,

Diego T. Terasaka,

Luiz E. R. Martins

et al.

Published: Nov. 17, 2024

The identification of bird species enables the creation machine learning models that can be employed for non-invasive monitoring populations. In this study, we present an advancement in assisted automated a training set classification species, with specific focus on Pantanal. Typically, process is conducted manually, which highly time-consuming approach. phase, propose comprehensive comparative testing to ascertain optimal methodologies feature extraction and clustering. Five clustering methods four were subjected testing. results our experiments demonstrate method purpose work was hierarchical clustering, using BirdNET extraction. This combination provided superior performance classifying construction sets.

Language: Английский

Citations

0