Link Prediction on a Community Graph: Common Neighbours, Adamic-Adar and Spectral Embedding AUC
Hide a fraction of the edges of a two-community stochastic block model and rank all unconnected pairs with common neighbours, Adamic-Adar and a rank-two spectral embedding. Change the community strength, density and hidden fraction to see AUC rise from chance to near perfect, with closed-form expected common-neighbour counts, a rank-sum versus ROC-area check and eigenvector residuals built in. A Blender-built data bench with the graph.
This simulator builds a random community-structured graph, gives every node a random 3D embedding, and hides a fraction of the real edges as a test set. A GCN-style neighbor-mean aggregation step — the same mechanism behind GCN, GraphSAGE and GAT-style message passing — repeatedly pulls each node's embedding toward the average of its visible (train) neighbors. Score every non-train pair by cosine similarity or raw dot product, sweep a threshold to see which held-out links get correctly predicted versus missed, watch false-positive predictions light up, and track a live AUC estimate as the embeddings converge.
Hide a fraction of the edges of a two-community stochastic block model and rank all unconnected pairs with common neighbours, Adamic-Adar and a rank-two spectral embedding. Change the community strength, density and hidden fraction to see AUC rise from chance to near perfect, with closed-form expected common-neighbour counts, a rank-sum versus ROC-area check and eigenvector residuals built in. A Blender-built data bench with the graph.
3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install