Please enable JavaScript.
Coggle requires JavaScript to display documents.
Pajek - Coggle Diagram
Pajek
Partition
classification/ clustering of the vertices in the network
input
Constant Partition
selected dimension
Default ( size)
Random Partition
random (one/ two mode partition)
Binarize
binary partition (0-1)
Fuse Clusters
selected cluster number >>> (new)
Canonical Partition
canonical form
vertex 1 (class 1)
next vertex (class 2)
Canonical Partition
(Decreasing frequencies)
canonical form
cluster 1 ~ the old cluster with the highest frequency
cluster 2 ~the old cluster with the second highest frequency
Make Network
Generate Network from Partition
Network
Random Network determined degree of vertices
Undirected
degrees of vertices in undirected network
Input
input degrees of vertices
Output
output degrees of vertices
Mode Network
generate two mode network
First set - vertices (v1... vn)
Second set - Clusters ( c0... cn )
Copy to VertexID
Convert partition clusters to vertex identificators
Make Permutation
selected partition
Make Cluster
Transform partition to cluster
Vertices from Selected Clusters
Put vertices from selected clusters
in Partition to Cluster
Random Representatives of each Cluster
Randomly select one
vertex (representative) from each cluster in Partition
Store representatives
to Cluster
Use as pivots in Pivot
MDS
Make Hierarchy
Transform partition to hierarchy (nested or not)
Copy to Vector
Copy partition to vector (V [i] := C[i])
Count, Min-Max Vector
info
cluster frequencies
minimum and maximum vector value
Info
General information
Vertices sorting
class numbers
(ascending/ descending)
most important vertices
Frequency distribution of class numbers
Average, median & standard
deviation of class numbers
Partitions
Operations on two partitions
(before)
Extract Sub Partition (Second from First)
partition that saves information about vertices (gender)
Add (First+Second)
two partitions
combining
Input and Output neighbors in acyclic networks
Min (First, Second)
Minimum of two partitions
Max (First, Second)
Maximum of two partitions
Fuse Partitions
add second to the end of the first
(2-mode networks )
Expand Partition
higher (original) dimension
First according to Second (Shrink)
first partition according
to shrinking determined by second partition
Insert First into Second according to Third (Extract)
current
partition
extracting selected classes defined by the
second partition from the first partition
modified sub-partition
the first partition
Intersection of Partitions
selected partitions
Cover with
p= partition
b= binary partition
c= selected cluster
q= number new partition
b(v) = 0
q(v) = p(v)
q(v) = c
Merge Partitions
p & q = partitions
b= binary partition
s= new partition
b(v) = 0 s(v) = p(v)
s(v) = q(v)
Make Random Network
input degrees
first & output degrees by the second partition
Functional Composition First*Second
f &g= two partitions
r= new partition
r[v] = (f *g)[v] = g[f[v]]
Info
Bivariate statistical measures between selected partitions
Cramer’s V, Rajski, Adjusted Rand Index
report contingency
table
compute Cramer’s V
Rajski coefficients
Adjusted Rand
Index
Spearman Rank
correlation coefficient
Vectors
(two vectors operation)
same dimension
scalar
Add (First+Second)
sum of selected vectors
Subtract (First-Second)
difference of selected vectors
Multiply (First*Second)
product of selected vectors
Divide (First/Second)
division of selected vectors
Min (First, Second)
smaller elements in selected vectors
Max (First, Second)
bigger elements in selected vectors
Fuse Vectors
fusion of vectors
Linear Regression
fit the two vectors using linear regression
Results
regression line
linear estimates of second vector
corresponding errors
Transform
two vectors to another two vectors
Cartesian
(Polar)
First vector --- x-coordinates
Second---y-coordinates
Results
vector containing polar radius
vector
containing polar angles in degrees
Polar
(Cartesian)
First vector --- polar radius
Second---polar angles in degrees
Results
vector containing x-coordinates
vector containing y-coordinates
Missing Values
values larger than 999.999.997
valid values or as missing values ?
Info
Pearson correlation coefficient between selected vectors.
Permutation
input
Create
Identity Permutation
selected dimension
Default dimension
(selected network size)
Random Permutation
selected dimension
Default dimension
(selected network size)
Inverse Permutation
selected permutation
Mirror Permutation
selected permutation
(sort in opposite direction)
Make Partition
Given Number of Clusters
given number
of clusters
Orbits
orbits in permutation
Copy to Vector
Copy permutation to vector
Info
Check if permutation is valid and return number of orbits
a renumbering of
vertices
Vector
(Operations)
Create
Constant Vector
selected dimension
Default dimension
(size of selected network )
Scalar
given Vector
Make
Partition
Intervals
selected dividing numbers in vector vertices
get appropriate class numbers
First Threshold and Step
Select first threshold and step in
which to increase threshold.
Selected Thresholds
Select all thresholds or number of classes
(#) in advance.
Copy to Partition by Truncating (Abs)
partition is absolute and
truncated vector
Permutation
Convert vector to permutation - sorting permutation
Cluster
Convert vector to cluster – select vertices with vector values
lower/higher than selected value.
2-Mode Network
Convert vector to 2-mode network (row or col).
Transform
Add Constant
vector values
Absolute values
elements
Absolute + Sqrt
square root of absolute components.
Multiply
constant
Truncate
truncated vector
Exp
exponential of vector
Ln
natural logarithm of vector
Power
selected power of vector
Normalize
Sum
normalize --- sum of elements = 1.
Max
normalize --- the maximum element = value 1.
Standardize
normalize---arithmetic mean=0 and
standard deviation= 1
Invert
inverse values of vector
Cumulatives
new vector u with cumulatives of vector v
u1 = v1; ui = ui1 + vi; i > 1
Missing Values
Select how values larger than 999.999.99
valid values or as missing values?
Info
General information
Vertices sorting
values
average
median
standard deviation
frequency distribution of
vector values
data object assigning a numerical value to
each vertex in a network