A hierarchical approach to improve the ant colony optimization algorithm

<p>The ant colony optimization algorithm (ACO) is a fast heuristic-based method for finding favorable solutions to the traveling salesman problem (TSP). When the data set reaches larger values however, the ACO runtime increases dramatically. As a result, clustering nodes into groups is an effective way to reduce the size of the problem while leveraging the advantages of the ACO algorithm. The method for recombining groups of nodes is explored by treating the graph as a hierarchy of clusters, and modifying the original ACO heuristic to operate on a hypergraph. This method of using hierarchical clustering is significantly faster than the original ACO algorithm, even when normal clustering techniques are applied, while producing improved tour lengths.</p>

Name thesis:degree_name

Master of Science (MS) in Computer Science

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science and Electrical Engineering

Year

2023

Author dc:creator

• Fischer, Bryan J.

• Numerical Analysis and Computation
• Theory and Algorithms

Statement dc:rights

• Access is available to all users

Repository record dc:identifier

https://dc.ewu.edu/theses/845

OAI identifier oai:identifier

oai:dc.ewu.edu:theses-1845