Table 1 Experimental results on the ACE2005 data set (%).
From: Joint event extraction model based on dynamic attention matching and graph attention networks
Approach | Trigger Identification (%) | Trigger Classification (%) | Argument Identification (%) | Argument Role(%) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
P | R | \({F_1}\) | P | R | \({F_1}\) | P | R | \({F_1}\) | P | R | \({F_1}\) | |
JointBeam | 74.7 | 62.8 | 68.2 | 71.4 | 60.3 | 65.4 | 66.9 | 45.2 | 53.9 | 61.2 | 41.5 | 49.5 |
DMCNN | 78.8 | 65.1 | 71.3 | 73.5 | 61.2 | 66.8 | 65.8 | 48.7 | 56.0 | 59.9 | 43.1 | 50.1 |
JRNN | 66.3 | 72.9 | 69.4 | 64.1 | 71.0 | 67.4 | 58.7 | 61.4 | 60.0 | 51.5 | 54.3 | 52.9 |
dbRNN | N/A | 71.7 | 67.4 | 69.5 | 69.7 | 63.0 | 66.2 | 64.1 | 50.1 | 56.2 | ||
JMEE | 78.5 | 70.1 | 74.1 | 73.9 | 70.1 | 71.9 | 69.7 | 63.8 | 66.6 | 64.5 | 52.1 | 57.6 |
GREE | N/A | 68.6 | 71.2 | 69.9 | 68.2 | 63.1 | 65.6 | 64.0 | 51.1 | 56.8 | ||
Our model | 77.8 | 70.5 | 74.0 | 76.3 | 72.2 | 74.2 | 70.2 | 64.0 | 67.0 | 65.2 | 52.8 | 58.3 |
