Skip to main content

Table 7 Results of various approaches considering performance for both identifying doubt and overall doubt classification (P, R, F denotes Precision, Recall, and F1 score, respectively, with P-y indicating Precision in identifying doubt or label ‘y’ during testing)

From: Automated doubt identification from informal reflections through hybrid sentic patterns and machine learning approach

Approach P-y R-y F-y P R F
doc2vec_CNN-LSTM 0.41 0.45 0.43 0.62 0.62 0.62
glove_LSTM 0.37 0.50 0.42 0.59 0.61 0.60
Hybrid (doc2vec_CNN-LSTM + Sentic pattern) 0.42 0.75 0.54 0.66 0.71 0.65
Hybrid (glove_LSTM + Sentic pattern) 0.40 0.82 0.54 0.66 0.71 0.63