Rumor Detection Based on Knowledge Enhancement and Graph Attention Network

Table 5

GAT results based on two different parsers, unlabeled attachment score, and labeled attachment score are indicators for evaluating parsers. The higher the score, the better the performance.

Parser	Performance		Dataset twitter
	UAS	LAS
Stanford	94.10	91.49	0.885
Biaffine	95.74	94.08	0.853