Table 8.

Performance measures of the text + image models on fakeddit dataset

Text+Image model	Fusion method	Validation	Test	Precision	Recall	F1-score
		accuracy	accuracy
BERT+Xception	Maximum	91.61%	91.87%	93.43%	93.07%	93.25%
BERT+Xception	Concatenate	91.67%	91.88%	93.31%	93.22%	93.26%
(BERT+Dense)+Xception	Maximum	91.68%	91.94%	93.76%	92.83%	93.29%
(BERT+Dense)+Xception	Concatenate	91.70%	91.87%	93.39%	93.29%	93.25%

Bold indicates models with better performance measures (here validation and Test accuracy)