CAREA: Cotraining Attribute and Relation Embeddings for Cross-Lingual Entity Alignment in Knowledge Graphs

Table 2

Comparison with the baseline methods.

Method
			MRR			MRR			MRR
^MTransE	^30.8%	^61.4%	^0.364	^27.9%	^57.5%	^0.349	^24.4%	^55.6%	^0.335
^JAPE	^41.2%	^74.5%	^0.490	^36.3%	^68.5%	^0.476	^32.4%	^66.7%	^0.430
^GCN-Align	^41.3%	^74.4%	^0.549	^39.9%	^74.5%	^0.546	^37.3%	^74.5%	^0.532
^MuGNN	^49.4%	^84.4%	^0.611	^50.1%	^85.7%	^0.621	^49.5%	^87.0%	^0.621
^BootEA	^62.9%	^84.8%	^0.703	^62.2%	^85.4%	^0.701	^65.3%	^87.4%	^0.731
^NAEA	^65.0%	^86.7%	^0.720	^64.1%	^87.3%	^0.718	^67.3%	^89.4%	^0.752
^CAREA-a	^22.1%	^51.8%	^0.320	^18.4%	^46.4%	^0.277	^16.9%	^45.4%	^0.264
^CAREA-s	^58.2%	^87.5%	^0.685	^58.5%	^87.8%	^0.689	^60.5%	^90.5%	^0.711
^CAREA	^69.8%	^90.6%	^0.772	^68.9%	^90.4%	^0.766	^71.3%	^92.5%	^0.789