Crate graph_neighbor_matching 

A graph similarity score using neighbor matching according to this paper.

TODO: Introduce EdgeWeight trait to abstract edge weight similarity.

Modules

graph

Structs

IgnoreNodeColors

Use IgnoreNodeColors to ignore node colors.

SimilarityMatrix

WeightedNodeColors

Use WeightedNodeColors to use the distance of both nodes node_color_weight as a measure of the similarity between these nodes.

Enums

ScoreNorm

Traits

Edges

Abstract representation of the edges of a node. Used by the algorithm.

Graph

Abstract representation of a Graph. Used by the algorithm.

NodeColorMatching

Describes the closeness of two nodes based on their colors. The color can be thought of as a node type. For example if you have a graph with nodes of type A and B which represent completely different things, you’d assign them a node_color_matching value of 0.0. This will tell our algorithm not to try to match these two nodes and their edges.

NodeColorWeight

The weight of the node color.

Functions

similarity_max_degree

similarity_min_degree

Type Aliases

EdgeWeight

The weight of an edge.