All MCQs

Visualization maps data to visual properties so relationships, trends, and anomalies can be perceived. It supports both analytical discovery and communication to an audience.

Jaccard similarity measures shared neighbors relative to the total distinct neighbors across both users. It ranges from zero for no overlap to one for identical sets.

An ego network describes the local social environment surrounding one selected entity. It is useful for studying neighborhood structure and individual network position.

A bridge is the sole connection linking two otherwise separated parts of an undirected graph. Removing it disconnects those parts from each other.

Spectral methods use eigenstructure of matrices such as the graph Laplacian to reveal global connectivity patterns. A selected eigenvector can guide a cut or lower-dimensional graph embedding.

Minimizing cut size alone can favor an unhelpful tiny partition separated from the rest. Balance constraints encourage groups of practical or comparable size.

Cut edges cross the boundary created by a partition. Many partitioning objectives try to reduce their number or total weight.

Graph partitioning assigns vertices to subsets according to an objective such as minimizing cross-group edges. It is related to, but not always identical with, community detection.

A bridge between otherwise dense regions carries a large fraction of cross-region shortest paths. Removing high-betweenness edges can therefore separate communities in divisive algorithms.

Local label updates allow densely connected regions to converge toward shared community identifiers. The method directly discovers groups without requiring labeled training examples.

Modularity compares observed within-community edges with an expected baseline that preserves aspects such as degree. Higher modularity suggests stronger community structure under that comparison.

Community boundaries reflect relatively stronger internal connectivity, not total separation. Connected components are separated absolutely because no path crosses between them.