[Retracted] Music Emotion Classification Method Based on Deep Learning and Explicit Sparse Attention Network

Table 3

Classification accuracy under different data preprocessing methods.

Preprocessing method	Happy	Sad	Relax	Anger	Average
Fine-grained segmentation	0.687	0.659	0.636	0.618	0.650
Vocal separation	0.678	0.660	0.643	0.632	0.653
Fine-grained segmentation + vocal separation	0.712	0.689	0.661	0.654	0.679
Proposed preprocessing method	0.737	0.723	0.698	0.688	0.712