Skip to main content
. 2025 Jan 23;11:e2595. doi: 10.7717/peerj-cs.2595

Table 3. Performance (in percentage) for entity prediction task on ICEWS05-15 and GDELT with time-aware metrics.

The best performance is highlighted in bold.

Model ICEWS05-15 GDELT
MRR H@1 H@3 H@10 MRR H@1 H@3 H@10
RGCRN (2018) 44.56 34.16 50.06 64.51 19.37 12.24 20.57 33.32
RE-NET (2020) 43.67 33.55 48.83 62.72 19.55 12.38 20.80 34.00
CyGNet (2021) 40.42 29.44 46.06 61.60 20.22 12.35 21.66 35.82
xERTE (2021) 46.62 37.84 52.31 63.92 19.45 11.92 20.84 34.18
TITer (2021) 47.60 38.29 52.74 64.86 18.19 11.52 19.20 31.00
RE-GCN (2021) 48.03 37.33 53.90 68.51 19.69 12.46 20.93 33.81
CEN (2022) 46.84 36.38 52.45 67.01 20.39 12.96 21.77 34.97
TiRGN (2022) 50.04 39.25 56.13 70.71 21.67 13.63 23.27 37.60
CENET (2023) 39.10 29.02 43.81 58.43 20.23 12.69 21.70 34.92
The proposed TRCL 50.12 39.08 56.39 70.87 21.85 13.68 23.55 38.10