Table 3 Performance comparison of different attention mechanisms on the drug–target prediction task.

From: Heterogeneous network drug-target interaction prediction model based on graph wavelet transform and multi-level contrastive learning

Attention Mechanism

AUC

AUPR

Semantic Attention

0.966 ± 0.016

0.888 ± 0.018

Similarity-Based Attention

0.955 ± 0.013

0.883 ± 0.006

Gated Attention

0.958 ± 0.017

0.884 ± 0.015

Multi-Head Attention

0.958 ± 0.009

0.883 ± 0.004

Scaled Dot-Product Attention

0.959 ± 0.015

0.885 ± 0.012

Back to article page