Phonetics and Ambiguity Comprehension Gated Attention Network for Humor Recognition

Table 2

Experimental results on Puns dataset. Best results are in bold. The results with superscript are imported from the literature [3, 15–17].

Models	Acc (%)	P (%)	R (%)	F1 (%)
SVM	83.85	85.91	82.52	84.18
HCFW2V	85.4	83.4	88.8	85.9
Syntactic	—	—	—	—
TM	74.5	75.2	72.3	73.7
CNN	86.1	86.4	86.4	85.7
CNN + HN + F	89.4	86.6	94.0	90.1
Bi-lstm + CNN	85.38	81.42	91.97	86.37
Bi-GRU	87.72	84.23	92.46	88.15
Bi-GRU + F	87.14	89.87	83.34	86.48
PACGA	88.69	88.94	92.76	90.81