Skip to main content
. 2024 Feb 20;10:e1831. doi: 10.7717/peerj-cs.1831

Table 8. Experiment results of FastText Embedding model including various N-grams with deep learning algorithms for ACPs250 dataset.

The best accuracy percentages are represented in bold letters.

MODELS ACC SEN SPE MCC AUC
FT(2)+CNN 88.99 86.27 89.75 77.63 93.51
FT(3)+CNN 83.99 80.58 87.42 67.29 90.20
FT(4)+CNN 57.99 55.79 59.83 15.74 60.29
FT(2)+LSTM 72.00 65.22 78.13 40.85 76.68
FT(3)+LSTM 75.99 70.34 82.49 49.32 82.17
FT(4)+LSTM 60.22 55.79 59.83 15.74 60.29
FT(2)+BiLSTM 82.99 79.68 86.34 70.01 90.01
FT(3)+BiLSTM 92.50 96.29 82.60 80.27 89.45
FT(4)+BiLSTM 66.15 55.79 59.83 15.74 60.29