Skip to main content
. 2019 Aug 28;19(17):3728. doi: 10.3390/s19173728

Table 8.

Results of FudanNLP in 17 classes for 4117 texts by SVM.

Model Accuracy and F1-Macro of Different Dimensions
200 300 500 800
Acc F1 Acc F1 Acc F1 Acc F1
FW 0.7931 0.5369 0.8037 0.5552 0.8091 0.5583 0.8156 0.5856
P2 0.5317 0.3603 0.5486 0.3662 0.5563 0.3672 0.5108 0.3524
PW 0.8012 0.4436 0.8002 0.4431 0.8011 0.4437 0.8005 0.4434
FP2 0.7931 0.5356 0.804 0.5355 0.8094 0.5707 0.8152 0.5815
FPW 0.8323 0.6327 0.8396 0.6465 0.8432 0.6334 0.8493 0.6494
FPC 0.8352 0.6081 0.8425 0.6230 0.8461 0.6376 0.8491 0.6629
Accuracy and F1-Macro of Baselines
Model Acc F1
TF–IDF 0.7628 0.6325
LDA 0.7346 0.432
Word2Vec 0.828 0.606