Table 2. Performance evaluation.

Performance of different trained classifiers on Waseem-dataset and Davidson-dataset in terms of F1-measure are reported in a and b, respectively.

Model	F1-Measure
Waseem and Hovy [2]	75
Waseem et al. [17]	80
Zhang et al. [29]	82
Park et al. [43]	83
BERTBASE	81
BERTBASE + Nonlinear Layers	76
BERTBASE + bi-LSTM	86
BERTBASE + CNN	88
(a) Performance evaluation on Waseem-dataset.
Model	F1-Measure
Davidson et al. [4]	84
Zhang et al. [29]	94
Waseem et al. [17]	89
MacAvaney et al. [13]	77
BERTBASE	91
BERTBASE + Nonlinear Layers	87
BERTBASE + bi-LSTM	92
BERTBASE + CNN	92
(b) Performance evaluation on Davidson-dataset.