Skip to main content

Table 2 Examples CDFs

From: Learning an enriched representation from unlabeled data for protein-protein interaction extraction

CDFs

Chi-square

CDFs

Chi-square

P2_Left=with

P1_Right=with

Inner_Area=with

Left_Area=prot0

-1_From_P2=with

0_From_P2=with prot2

2_From_P1=with

Inner_Area=,

2_From_P1=- prot2

-1_From_P2=prot1 -

219.34

194.41

183.05

168.05

118.53

118.53

106.68

106.62

93.10

93.10

2_From_P1=prot1 - prot2

0_From_P2=prot1 - prot2

P1_Right=prot0

Inner_Area=prot0

P1_Right=interacts with

-1_From_P2=between prot1 and

P2_Left=between

-3_From_P2=between

-2_From_P2=between prot1

P1_Left=binds

93.10

93.10

83.61

83.09

80.94

79.3

75.91

74.4

74.4

68.17

  1. Since the CDFs extracted from the training set of each round of ten-fold cross validation are highly overlapped, here we just list the top results of the first round. “Chi-square” is the score given by Chi-square feature selection method.