Title :
Hybrid local search approximation algorithm for solving the capacitated Max-K-cut problem
Author :
Alqallaf, Safaa ; Almulla, Mohammed ; Niepel, Ludovit ; Newborn, Monty
Author_Institution :
Comput. Sci. Dept., Kuwait Univ., Safat, Kuwait
Abstract :
In this article we propose a new hybrid local search approximation algorithm for solving the capacitated Max-k-cut problem and contrast its performance with two local search approximation algorithms. The first of which uses a swapping neighborhood search technique, whereas the second algorithm uses a vertex movement method. We analyze the behavior of the three algorithms with respect to running time complexity, number of iterations performed and the total weight sum of the cut edges. The experimental results show that our proposed hybrid algorithm outperforms its rivals at all levels.
Keywords :
computational complexity; search problems; capacitated max-k-cut problem; hybrid local search approximation algorithm; swapping neighborhood search technique; time complexity; vertex movement method; Algorithm design and analysis; Approximation algorithms; Approximation methods; Partitioning algorithms; Search problems; Silicon; Time complexity; Capacitated Max-k-cut problem; Local search approximation algorithms; Swapping neighborhood; Vertex Movement;
Conference_Titel :
Web Applications and Networking (WSWAN), 2015 2nd World Symposium on
Conference_Location :
Sousse
Print_ISBN :
978-1-4799-8171-7
DOI :
10.1109/WSWAN.2015.7210302