[Retracted] A Deep Learning Approach for a Source Code Detection Model Using Self-Attention

Table 4

Clone detection models with various self-attention layers (percentage).

Description	F1 score (BigCloneBench)	F1 score (OJClone)
No self-attention	89.6	87.9
Self-attention layer with no parameters	92.4	91.7
At-BiLSTM	94.3	95.8