1. Holistic Result

Model

ctb

ckip

sent_long

97.59

96.21

sent_1

97.68

96.23

2. Break-down Performance

Model Setting: sent_long

Dataset: ctb

MF-et

MF-tt

F-ent

F-tok

R-eLen

R-sLen

R-oov

R-tag

0:(0,) 1:(0,0.5) 2:(0.5,1) 3:(1,)

0:(0,) 1:(0,0.5) 2:(0.5,1) 3:(1,)

0:(0,) 1:(0.001,0.009) 2:(0.01,0.107) 3:(0.109,1)

0:(0,) 1:(0.001,0.026) 2:(0.027,0.08) 3:(0.081,1)

0:(1,) 1:(2,) 2:(3,) 3:(4,11)

0:(2,42) 1:(43,62) 2:(63,91) 3:(92,246)

0:(0,) 1:(0.0041,0.0172) 2:(0.0175,0.0312) 3:(0.0323,1)

0:('BE',) 1:('BME',) 2:('BMME',) 3:('S',)

Dataset: ckip

MF-et

MF-tt

F-ent

F-tok

R-eLen

R-sLen

R-oov

R-tag

0:(0,) 1:(0,0.5) 2:(0.5,1) 3:(1,)

0:(0,) 1:(0,0.5) 2:(0.5,1) 3:(1,)

0:(0,) 1:(0.001,0.007) 2:(0.008,0.121) 3:(0.123,1)

0:(0,) 1:(0.001,0.02) 2:(0.021,0.063) 3:(0.064,1)

0:(1,) 1:(2,) 2:(3,) 3:(4,11)

0:(2,9) 1:(10,13) 2:(14,19) 3:(20,80)

0:(0,) 1:(0.0213,0.0667) 2:(0.0714,0.143) 3:(0.154,0.667)

0:('BE',) 1:('BME',) 2:('BMME',) 3:('S',)

Model Setting: sent_1

Dataset: ctb

MF-et

MF-tt

F-ent

F-tok

R-eLen

R-sLen

R-oov

R-tag

0:(0,) 1:(0,0.5) 2:(0.5,1) 3:(1,)

0:(0,) 1:(0,0.5) 2:(0.5,1) 3:(1,)

0:(0,) 1:(0.001,0.009) 2:(0.01,0.107) 3:(0.109,1)

0:(0,) 1:(0.001,0.026) 2:(0.027,0.08) 3:(0.081,1)

0:(1,) 1:(2,) 2:(3,) 3:(4,11)

0:(2,42) 1:(43,62) 2:(63,91) 3:(92,246)

0:(0,) 1:(0.0041,0.0172) 2:(0.0175,0.0312) 3:(0.0323,1)

0:('BE',) 1:('BME',) 2:('BMME',) 3:('S',)

Dataset: ckip

MF-et

MF-tt

F-ent

F-tok

R-eLen

R-sLen

R-oov

R-tag

0:(0,) 1:(0,0.5) 2:(0.5,1) 3:(1,)

0:(0,) 1:(0,0.5) 2:(0.5,1) 3:(1,)

0:(0,) 1:(0.001,0.007) 2:(0.008,0.121) 3:(0.123,1)

0:(0,) 1:(0.001,0.02) 2:(0.021,0.063) 3:(0.064,1)

0:(1,) 1:(2,) 2:(3,) 3:(4,11)

0:(2,9) 1:(10,13) 2:(14,19) 3:(20,80)

0:(0,) 1:(0.0213,0.0667) 2:(0.0714,0.143) 3:(0.154,0.667)

0:('BE',) 1:('BME',) 2:('BMME',) 3:('S',)

3. Self-diagnosis

Model Setting: sent_long

ctb

ckip

Model Setting: sent_1

ctb

ckip

4. Aided-diagnosis

Model Setting: sent_long - sent_1

ctb

ckip

5. Heatmap

Model Setting: sent_long - sent_1

ctb

ckip