Classes related to graph clustering.
Class 

Class representing a clustering of an arbitrary ordered set. 
Class 

The cohesive block structure of a graph. 
Class 

Class representing a cover of an arbitrary ordered set. 
Class 

The hierarchical clustering (dendrogram) of some dataset. 
Class 

The clustering of the vertex set of a graph. 
Class 

The cover of the vertex set of a graph. 
Class 

The dendrogram resulting from the hierarchical clustering of the vertex set of a graph. 
Function  compare 
Compares two community structures using various distance measures. 
Function  split 
Calculates the splitjoin distance between two community structures. 
Function  _handle 
Handles the mark_groups=... keyword argument in plotting methods of clusterings. 
Function  _prepare 
Auxiliary method that takes two community structures either as membership lists or instances of Clustering , and returns a tuple whose two elements are membership lists. 
Compares two community structures using various distance measures.
Parameters  
comm1  the first community structure as a membership list or as a Clustering object. 
comm2  the second community structure as a membership list or as a Clustering object. 
method  the measure to use. "vi" or "meila" means the variation of information metric of Meila (2003), "nmi" or "danon" means the normalized mutual information as defined by Danon et al (2005), "splitjoin" means the splitjoin distance of van Dongen (2000), "rand" means the Rand index of Rand (1971), "adjusted_rand" means the adjusted Rand index of Hubert and Arabie (1985). 
remove  whether to remove None entries from the membership lists. This is handy if your Clustering object was constructed using VertexClustering.FromAttribute using an attribute which was not defined for all the vertices. If remove_none is False, a None entry in either comm1 or comm2 will result in an exception. If remove_none is True, None values are filtered away and only the remaining lists are compared. 
Returns  
the calculated measure.  
Unknown Field: newfield  
ref  Reference 
Unknown Field: ref  
Meila M: Comparing clusterings by the variation of information. In: Scholkopf B, Warmuth MK (eds). Learning Theory and Kernel Machines: 16th Annual Conference on Computational Learning Theory and 7th Kernel Workship, COLT/Kernel 2003, Washington, DC, USA. Lecture Notes in Computer Science, vol. 2777, Springer, 2003. ISBN: 9783540407201.  
Danon L, DiazGuilera A, Duch J, Arenas A: Comparing community structure identification. J Stat Mech P09008, 2005.  
van Dongen D: Performance criteria for graph clustering and Markov cluster experiments. Technical Report INSR0012, National Research Institute for Mathematics and Computer Science in the Netherlands, Amsterdam, May 2000.  
Rand WM: Objective criteria for the evaluation of clustering methods. J Am Stat Assoc 66(336):846850, 1971.  
Hubert L and Arabie P: Comparing partitions. Journal of Classification 2:193218, 1985. 
Calculates the splitjoin distance between two community structures.
The splitjoin distance is a distance measure defined on the space of partitions of a given set. It is the sum of the projection distance of one partition from the other and vice versa, where the projection number of A from B is if calculated as follows:
 For each set in A, find the set in B with which it has the maximal overlap, and take note of the size of the overlap.
 Take the sum of the maximal overlap sizes for each set in A.
 Subtract the sum from n, the number of elements in the partition.
Note that the projection distance is asymmetric, that's why it has to be calculated in both directions and then added together. This function returns the projection distance of comm1 from comm2 and the projection distance of comm2 from comm1, and returns them in a pair. The actual splitjoin distance is the sum of the two distances. The reason why it is presented this way is that one of the elements being zero then implies that one of the partitions is a subpartition of the other (and if it is close to zero, then one of the partitions is close to being a subpartition of the other).
Parameters  
comm1  the first community structure as a membership list or as a Clustering object. 
comm2  the second community structure as a membership list or as a Clustering object. 
remove  whether to remove None entries from the membership lists. This is handy if your Clustering object was constructed using VertexClustering.FromAttribute using an attribute which was not defined for all the vertices. If remove_none is False, a None entry in either comm1 or comm2 will result in an exception. If remove_none is True, None values are filtered away and only the remaining lists are compared. 
Returns  
the projection distance of comm1 from comm2 and vice versa in a tuple. The splitjoin distance is the sum of the two.  
See Also  
compare_communities() with method = "splitjoin" if you are not interested in the individual projection distances but only the sum of them.  
Unknown Field: newfield  
ref  Reference 
Unknown Field: ref  
van Dongen D: Performance criteria for graph clustering and Markov cluster experiments. Technical Report INSR0012, National Research Institute for Mathematics and Computer Science in the Netherlands, Amsterdam, May 2000. 
Handles the mark_groups=... keyword argument in plotting methods of clusterings.
This is an internal method, you shouldn't need to mess around with it. Its purpose is to handle the extended semantics of the mark_groups=... keyword argument in the __plot__ method of VertexClustering
and VertexCover
instances, namely the feature that numeric IDs are resolved to clusters automatically.
Auxiliary method that takes two community structures either as membership lists or instances of Clustering
, and returns a tuple whose two elements are membership lists.
This is used by compare_communities
and split_join_distance
.
Parameters  
comm1  the first community structure as a membership list or as a Clustering object. 
comm2  the second community structure as a membership list or as a Clustering object. 
remove  whether to remove None entries from the membership lists. If remove_none is False, a None entry in either comm1 or comm2 will result in an exception. If remove_none is True, None values are filtered away and only the remaining lists are compared. 