Table 4 The comparison of different systems on the CovSI test set
From: Multi-probe attention neural network for COVID-19 semantic indexing
Model | EBP (%) | EBR (%) | EBF (%) | MaP (%) | MaR (%) | MaF (%) | MiP (%) | MiR (%) | MiF (%) |
|---|---|---|---|---|---|---|---|---|---|
MPANN | 87.41 | 63.52 | 71.20 | 97.03 | 50.44 | 55.02 | 88.62 | 62.78 | 73.49 |
BioTrans [30] | 87.02 | 62.74 | 70.47 | 97.17 | 47.63 | 52.23 | 87.99 | 61.92 | 72.68 |
AttentionMeSH [29] | 81.18 | 54.52 | 63.08 | 88.51 | 51.15 | 54.36 | 81.57 | 53.48 | 64.60 |
FullMeSH [28] | 88.40 | 51.92 | 63.29 | 95.11 | 57.56 | 60.47 | 88.44 | 51.92 | 65.43 |
MeSHProbeNet-P [26] | 82.81 | 54.36 | 65.64 | 95.64 | 57.66 | 61.29 | 83.33 | 57.14 | 67.79 |