Table 5.

Results of FudanNLP in three classes for 1500 texts by KNN.

Model	Accuracy and F1-Macro of Different Dimensions
	200		300		500		800
	Acc	F1	Acc	F1	Acc	F1	Acc	F1
FW	0.94	0.9383	0.9353	0.934	0.93	0.9267	0.9253	0.9221
P2	0.9127	0.9119	0.9227	0.9214	0.9147	0.9139	0.9153	0.9145
PW	0.96	0.9598	0.9527	0.9524	0.9607	0.9605	0.96	0.96
FP2	0.934	0.9383	0.934	0.9326	0.9307	0.9284	0.926	0.9228
FPW	0.9493	0.9489	0.96	0.9598	0.95	0.9496	0.9513	0.951
FPC	0.96	0.9482	0.9527	0.9605	0.9607	0.9516	0.96	0.9494
Accuracy and F1-Macro of Baselines
Model	Acc				F1
TF–IDF	0.5047				0.581
LDA	0.9586				0.9584
Word2Vec	0.9467				0.9457