A Semi-Supervised Framework for MMMs-Induced Fuzzy Co-Clustering with Virtual Samples

Table 6

Comparison of recognition rates with and without virtual objects (Cora-training).

Number of virtual objects per supervised object
	0.01	0.05	0.1	0.2
No virtual objects	0.528
1 (GBD)	0.539	0.540	0.548	0.536
1 (GBA)	0.539	0.546	0.536	0.537
2 (GBD)	0.540	0.536	0.537	0.544
2 (GBA)	0.539	0.541	0.543	0.542
2 (GBD+GBA)	0.542	0.541	0.530	0.541
4 (GBD+GBA)	0.528	0.529	0.531	0.530
5 (GBD)	0.525	0.512	0.529	0.536
5 (GBA)	0.525	0.521	0.503	0.504