Table 5.

The F1 score performance on simple events, nested events, and all events on the multilevel event extraction corpus.

Subtask and model			Simple (%)		Nested (%)		All (%)
Trigger
	CNNa	79.52		78.80		78.52
	RLb+KBsc	N/Ad		N/A		N/A
	DeepEventMine	N/A		79.12		N/A
	HANNe	N/A		N/A		N/A
	Our model	79.96f		80.05f		80.18f
Event
	CNN	61.33		54.29		58.87
	RL+KBs	N/A		58.69		60.09
	DeepEventMine	N/A		51.73		61.87
	HANN	77.08f		45.46		59.74
	Our model	64.85		61.26f		62.80f

^aCNN: convolutional neural network.

^bRL: reinforcement learning.

^cKB: knowledge base.

^dN/A: not applicable.

^eHANN: hierarchical artificial neural network.

^fThe best value compared with other models.