Documentation ¶
Overview ¶
Package graph contains code and logic for computing the [Markov cluster](https://micans.org/mcl/) from a graph of keywords with edges weighted by the similarity scores among them.
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type ClusterGroup ¶
ClusterGroup is a cluster of highly-related keywords with respect to the similarity of their SERP members
type Graph ¶
type Graph struct {
// contains filtered or unexported fields
}
Graph builds a network of keywords and their relationship to other keywords with respect to the similarity of their SERPs
func (Graph) FindClusters ¶
func (g Graph) FindClusters(kd rankings.KeywordData) ([]ClusterGroup, error)
FindClusters adds gathered SERP data to a graph, computes the RBO weights among all SERPs, and returns clusters of keywords whose SERP members are similar
type Option ¶
type Option func(g *Graph)
Option configures a graph
func WithClusterInflation ¶
WithClusterInflation configures the graph to use the specified inflation value to compute graph clusters
func WithClusterMaxIterations ¶
WithClusterMaxIterations configures the graph to limit the number of computation steps used when computing graph clusters
func WithClusterPower ¶
WithClusterPower configures the graph to use the specified power value to compute graph clusters
func WithRBOPValue ¶
WithRBOPValue configures the graph to use the specified RBO probability value