Skip to main content

Table 8 Average test accuracies in five-fold validation for the different subrelations

From: Incorporating representation learning and multihead attention to improve biomedical cross-sentence n-ary relation extraction

Method Ternary
  Single Cross
With drug-gene relation 83.1 89.5
With drug-mutation relation 82.7 89.5
With two relation 87.3* 91.9*