Figures & data
Fig. 1 Histogram of within-community degree distributions from three bipartite networks with size , obtained from (a) a simulation of a SBM, (b) a simulation of a DCSBM, and (c) a real-world computer network (ICL2, see Section 6.2).
![Fig. 1 Histogram of within-community degree distributions from three bipartite networks with size 439×60,635, obtained from (a) a simulation of a SBM, (b) a simulation of a DCSBM, and (c) a real-world computer network (ICL2, see Section 6.2).](/cms/asset/bdaf66fb-d14e-4ea8-b187-8d2301d66cac/utch_a_2008503_f0001_c.jpg)
Fig. 2 Scatterplots of the two-dimensional ASE of a simulated DCSBM with (a) K = 4, and (b) K = 2. also highlights the true and estimated latent position for six nodes, with the corresponding 50%, 75%, and 90% contours from the ASE-CLT, and the estimated latent positions for
from simulated DCSBM adjacency matrices
.
![Fig. 2 Scatterplots of the two-dimensional ASE of a simulated DCSBM with (a) K = 4, and (b) K = 2. Figure 2(b) also highlights the true and estimated latent position for six nodes, with the corresponding 50%, 75%, and 90% contours from the ASE-CLT, and the estimated latent positions x̂1,l for x1 from simulated DCSBM adjacency matrices Al, l=1,…,1000.](/cms/asset/2bd232e8-5ae9-46f5-85fb-58535d0d0ff9/utch_a_2008503_f0002_c.jpg)
Fig. 3 Plots of and
, obtained from the two-dimensional ASE of a simulated DCSBM. Joint (green) and community-specific (blue and red) marginal distributions with MLE Gaussian fit are also shown.
![Fig. 3 Plots of x̂i, x˜i=x̂i/||x̂i|| and θ̂i=f2(x̂i), obtained from the two-dimensional ASE of a simulated DCSBM. Joint (green) and community-specific (blue and red) marginal distributions with MLE Gaussian fit are also shown.](/cms/asset/f2fbe969-7c50-44c7-b534-f12e091490bd/utch_a_2008503_f0003_c.jpg)
Fig. 4 Boxplots for simulations of a degree-corrected stochastic blockmodel with
nodes, K = 3, equal number of nodes allocated to each group, and B described in (9), corrected by parameters ρi sampled from a
distribution.
![Fig. 4 Boxplots for N=1000 simulations of a degree-corrected stochastic blockmodel with n=2000 nodes, K = 3, equal number of nodes allocated to each group, and B described in (9), corrected by parameters ρi sampled from a Uniform(0,1) distribution.](/cms/asset/5bc687f8-08b1-4953-ad04-b6e66f7f0cc6/utch_a_2008503_f0004_c.jpg)
Table 1 Estimated performance for N = 250 simulated DCSBMs and bipartite DCScBMs.
Fig. 5 Estimated performance for N = 250 simulated DCSBMs and DCScBMs, for . For bipartite DCScBMs,
.
![Fig. 5 Estimated performance for N = 250 simulated DCSBMs and DCScBMs, for n∈{100,200,500,1000,2000}. For bipartite DCScBMs, n′∈{150,300,750,1500,3000}.](/cms/asset/336cea74-44f7-4dba-b7a3-69f6164376a3/utch_a_2008503_f0005_c.jpg)
Table 2 Summary statistics for the Imperial College London computer networks.
Table 3 Estimates of (d, K) and ARIs for the embeddings and
for
and alternative methodologies.