A Knowledge Representation Method for Question Answering Service in Mobile Edge Computing Environment

Table 4

Comparison result with neural network methods on FB15K-237 and WN18RR.

	FB15K−237					WN18RR
	MR	MRR	Hits@10	Hits@3	Hits@1	MR	MRR	Hits@10	Hits@3	Hits@1
CapsE	303	0.523	0.593	−	−	719	0.415	0.56	−	−
ConvKB	216	0.289	0.471	0.327	0.198	1295	0.265	0.558	0.445	0.058
ConvE	245	0.312	0.497	0.341	0.225	4464	0.456	0.531	0.47	0.419
SCAN	−	0.35	0.54	0.39	0.26	−	0.47	0.54	0.48	0.43
R-GCN	−	0.249	0.417	0.264	0.151	−	−	−	−	−
KRDGC	158	0.527	0.662	0.563	0.46	1850	0.424	0.584	0.489	0.34