From Daan
Jump to: navigation, search
(Paper)
 
(12 intermediate revisions by the same user not shown)
Line 1: Line 1:
  
 +
==Paper==
  
This is a temporary redirection page associated with a replication study of Van Horn et al for review transparency. We're currently finishing the paper.
+
Here is our [http://heuristieken.nl/resources/VanHornetal(2018)-aPredictiveDataAnalyticfortheHardnessofHamiltonianCycleInstances.pdf paper].
  
  
==Interative Graphs==
+
'''Corrections, additions & developments:'''
 +
 
 +
 
 +
CORRECTION: I misspelled the name of Edward Reingold as "Rheingold", possibly being confused by the Wagner opera.
  
The interactive graphs for Hamiltonian cycles can be found [https://hamiltoncycle.gijsvanhorn.nl/ here].
 
  
 +
ADDITION to page 93: "when recursing, always prioritize a higher degree vertex over a lower degree vertex.".  It should be understood "a higher degree vertex" means "a vertex which is not already in the path, that has the largest number of edges ''to unvisited vertices''". In other words: the priority list of available vertices is determined and sorted during each recursion call.
  
== Source Data for Interative Graphs==
+
==Interactive Graphs==
  
Cheeseman's algorithm on 16 nodes graphs: [https://hamiltoncycle.gijsvanhorn.nl/data/16-node-Cheeseman-results.dsv]
+
The interactive graphs for Hamiltonian cycles can be found [https://hamiltoncycle.gijsvanhorn.nl/ here].
Average and median computational costs: [https://hamiltoncycle.gijsvanhorn.nl/data/16-node-Cheeseman-derived-results.dsv]
 
  
Cheeseman's algorithm on 24 node graphs: [https://hamiltoncycle.gijsvanhorn.nl/data/24-node-Cheeseman-results.dsv]
 
Average and median computational costs: [https://hamiltoncycle.gijsvanhorn.nl/data/24-node-Cheeseman-derived-results.dsv]
 
  
Van Horn's algorithm on 16 node graphs: [https://hamiltoncycle.gijsvanhorn.nl/data/16-node-Inverse-Cheeseman-results.dsv]
 
Average and median computational costs: [https://hamiltoncycle.gijsvanhorn.nl/data/16-node-Inverse-Cheeseman-derived-results.dsv]
 
  
Van Horn's algorithm on 24 node graphs: [https://hamiltoncycle.gijsvanhorn.nl/data/24-node-Inverse-Cheeseman-results.dsv]
+
==Source Data for Interactive Graphs==
Average and median computational costs: [https://hamiltoncycle.gijsvanhorn.nl/data/24-node-Inverse-Cheeseman-derived-results.dsv]
 
  
Vandegriend and Culberson's algorithm on 16 node graphs: [https://hamiltoncycle.gijsvanhorn.nl/data/16-node-Pruning-results.dsv]
+
Download our [http://heuristieken.nl/resources/VanHornetal(2018)-results.xls results] here.
Average and median computational costs: [https://hamiltoncycle.gijsvanhorn.nl/data/16-node-Pruning-derived-results.dsv]
 
  
Vandegriend and Culberson's algorithm on 24 node graphs: [https://hamiltoncycle.gijsvanhorn.nl/data/24-node-Pruning-results.dsv]
 
Average and median computational costs [https://hamiltoncycle.gijsvanhorn.nl/data/24-node-Pruning-derived-results.dsv]
 
  
  
== Source Code for the algorithms==
+
==Source Code for the Algorithms==
  
And here is the [https://hamiltoncycle.gijsvanhorn.nl/sourcecode source code] of the algorithms.
+
And here is the [http://heuristieken.nl/resources/VanHornetal(2018)-sourcecode.zip source code] of the algorithms.

Latest revision as of 13:53, 15 June 2019

Paper

Here is our paper.


Corrections, additions & developments:


CORRECTION: I misspelled the name of Edward Reingold as "Rheingold", possibly being confused by the Wagner opera.


ADDITION to page 93: "when recursing, always prioritize a higher degree vertex over a lower degree vertex.". It should be understood "a higher degree vertex" means "a vertex which is not already in the path, that has the largest number of edges to unvisited vertices". In other words: the priority list of available vertices is determined and sorted during each recursion call.

Interactive Graphs

The interactive graphs for Hamiltonian cycles can be found here.


Source Data for Interactive Graphs

Download our results here.


Source Code for the Algorithms

And here is the source code of the algorithms.