Skip to main content

Table 3 Performance comparison with and without different sub-modules

From: RPf-GCNs: reciprocal perspective driven fused GCNs for rumor detection on social media

Modules

PHEME

SemEval

Macro-F1

Acc

Macro-F1

Acc

Combined

0.736

0.748

0.461

0.658

–Conv

0.578

0.579

0.295

0.403

\(({q \rightarrow p})\)

0.701

0.7

0.405

0.643

\(({p \rightarrow q})\)

0.421

0.471

0.381

0.547

–Dir

0.525

0.536

0.405

0.556

GCN

0.446

0.493

0.371

0.547