The correct answer is Edge thickness.

Edge thickness is the visual feature that encodes the weight or number of dependencies between components in a graph. Visualizers scale edge thickness so that thicker lines represent stronger or more numerous dependencies and thinner lines represent weaker or fewer dependencies. This makes thickness the most direct way to compare the relative weight of connections between specific components.

The option Node color is incorrect because color is usually used to show categories or a scalar metric such as risk level or status rather than the numeric weight of a specific dependency.

The option Edge length is incorrect because length most often reflects the layout algorithm or spatial separation and it is not a reliable encoding for dependency weight in typical risk graphs.

The option Node degree is incorrect because degree reports how many connections a node has overall and it does not show the weight of individual edges between pairs of components. Degree helps identify highly connected nodes but not the strength of each dependency.

Full AWS Practitioner Certification Question