Short Text Paraphrase Identification Model Based on RDN-MESIM

Table 2

Different models results on PI task using tax intelligence consulting question pairs’ data.

Model	Precision	Recall	Fmeasure	Accuracy (%)
DSSM [12]	0.8412	0.8187	0.8297	83.64
CDSSM [13]	0.9380	0.8814	0.9088	92.20
MVLSTM [16]	0.9940	0.9112	0.9508	95.94
ARCII [17]	0.9710	0.9463	0.9584	96.47
ABCNN [18]	0.9559	0.9344	0.9450	94.46
DIIN [22]	0.9763	0.9511	0.9635	96.10
DRCN [23]	0.9401	0.9079	0.9237	93.40
RDN-MESIM	0.9889	0.9612	0.9748	97.63