Fig. 1From: Neural side effect discovery from user credibility and experience-assessed online health discussionsThe neural architecture of our proposed NEAT. The wu and vu boxes denote Credibility Weight (CW) component and User Expertise (UE) component. The yellow boxes and blue boxes denote Cluster Attention (CA) component and neural text encoders with attention. The highlighted words in red denoted the text segments that are being attended by the encoder. The ×, \(\sum \), and σ symbols denote the multiplication, summation, and sigmoid, respectivelyBack to article page