Table 3 Shows the comparative experimental results of the performance of TLAM-EA against three other methods: RNN, Hybrid, Magellan, and FLAM-EA.
From: Contextual semantics graph attention network model for entity resolution
Models | Amazon-Google | BeerAdvo-RateBeer | ||||
|---|---|---|---|---|---|---|
P (%) | R (%) | F1 (%) | P (%) | R (%) | F1 (%) | |
RNN | 59.33 | 48.12 | 52.77 | 74.82 | 70.00 | 71.34 |
Hybrid | 58.82 | 64.02 | 60.51 | 73.44 | 70.00 | 71.08 |
Magellan | 67.7 | 38.5 | 49.1 | 68.4 | 92.9 | 78.8 |
TLAM-ER | 61.71 \(\pm\) 1.40 | 64.29 \(\pm\) 2.31 | 62.97 \(\pm\) 1.70 | 78.24 \(\pm\) 2.33 | 79.84 \(\pm\) 4.14 | 79.03 \(\pm\) 1.93 |
PBAL-EM | _ | _ | 42.40 | _ | _ | 86.70 |
CSGAT | 63.36 \(\pm\) 1.91 | 68.61 \(\pm\) 1.44 | 65.88 \(\pm\) 0.85 | 80.66 \(\pm\) 0.93 | 84.91 \(\pm\) 1.01 | 82.73 \(\pm\) 1.22 |
