MIT Press

Table 2.

Brief overview of the four algorithms’ properties relevant for topic reconstruction

	Community definition/separation-cohesion	Degree of local and global information usage	Hierarchies	Overlapping	Preference of size	Users’ degrees of freedom
Louvain	Dense community, focus on separation	Mainly global	Not directly	No	No	Resolution parameter, optimization function, random seed
Leiden	Dense community, focus on separation and slightly cohesion	Global and partly local	Not directly	No	No	Resolution parameter, optimization function, random seed
OSLOM	Statistically significant community, cohesion likely, separation not an issue	Local and partly global	Yes	Pervasive	No	Coverage parameter, p-value, random seed, singletons
Infomap	Nodes’ closeness, focus on separation	Mainly global	Yes	In boundaries	No	Two- or multilevel solution, detection of overlaps, random seed

	Community definition/separation-cohesion	Degree of local and global information usage	Hierarchies	Overlapping	Preference of size	Users’ degrees of freedom
Louvain	Dense community, focus on separation	Mainly global	Not directly	No	No	Resolution parameter, optimization function, random seed
Leiden	Dense community, focus on separation and slightly cohesion	Global and partly local	Not directly	No	No	Resolution parameter, optimization function, random seed
OSLOM	Statistically significant community, cohesion likely, separation not an issue	Local and partly global	Yes	Pervasive	No	Coverage parameter, p-value, random seed, singletons
Infomap	Nodes’ closeness, focus on separation	Mainly global	Yes	In boundaries	No	Two- or multilevel solution, detection of overlaps, random seed