edu.iu.iv.analysis.pathfindernetworkscaling
Class PFNET

java.lang.Object
  |
  +--edu.iu.iv.analysis.pathfindernetworkscaling.PFNET

public class PFNET

extends java.lang.Object

Author:

Team IVC

Constructor Summary

PFNET(int q, int r, DoubleMatrixModel matrix)
Initializes the PFNET algorithm.

Method Summary

void	applyScaling()
DoubleMatrixModel	getResultMatrix()

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

PFNET

public PFNET(int q,
             int r,
             DoubleMatrixModel matrix)
      throws java.lang.IllegalArgumentException

Initializes the PFNET algorithm. The q parameter must be at most n - 1 where n is the number of rows (or columns) of the matrix.
The r parameter must be non-negative. A value of 0 is taken to mean that the Minkowski metric is to be evaluated in the limit r -> infinity.
The input matrix must be a symmetric square matrix. The elements of the matrix are treated as weights of the edges and the rows (and columns) are treated as vertices of a graph. Since only symmetric matrices are considered, this means that this PFNET applies only to undirected graphs. The weights are considered as distance , hence a higher weight indicates higher distance. A weight of 0.0 is assumed on the diagonal since the weight from a vertex to itself is zero. However , a weight of 0.0 at a non-diagonal location is taken to mean that the corresponding edge does not exist in the graph and consequently, the weight is interpreted as being infinite . Hence, only the strictly upper triangular part of the matrix is significant.

Parameters:

q - The q parameter of the PFNET(r, q)

r - The r parameter of the PFNET(r, q)

matrix - The matrix to which the PFNET is to be applied.

Throws:

java.lang.IllegalArgumentException

Method Detail

applyScaling

public void applyScaling()

getResultMatrix

public DoubleMatrixModel getResultMatrix()