Multi-Label Classifier Chains for Bird Sound

Bird sound data collected with unattended microphones for automatic surveys, or mo-bile devices for citizen science, typically con-tain multiple simultaneously vocalizing birds of different species. However, few works have considered the multi-label structure in bird-song. We propose to use an ensemble of clas-sifier chains combined with a histogram-of-segments representation for multi-label clas-sification of birdsong. The proposed method is compared with binary relevance and three multi-instance multi-label learning (MIML) algorithms from prior work (which focus more on structure in the sound, and less on structure in the label sets). Experiments are conducted on two real-world birdsong datasets, and show that the proposed method usually outperforms binary relevance (using the same features and base-classifier), and is better in some cases and worse in others com-pared to the MIML algorithms. 1