Fig. 3
From: GraphTar: applying word2vec and graph neural networks to miRNA target prediction
A plot visualizing models performance on test datasets. Among these datasets, miTAR emerged as the top-performing method for the DeepMirTar dataset. On the miRAW dataset, the most effective approach was GraphTarGAT (abbreviated as GAT), while for the combined MirTarRaw dataset, the highest metric scores were achieved by miRAW. The abbreviations GCN and GRAPHSAGE correspond to GraphTarGCN and GraphTarSAGE, respectively