Grasp with path relinking for the sumcut problem

Jesús Sánchez-Oro, Abraham Duarte
January, 2012
Cite

Abstract

This paper proposes a GRASP algorithm combined with Path Relinking to solve the SumCut minimization problem. In the SumCut problem one is given a graph with n nodes and must label the nodes in a way that each node receives a unique label from the set1,2,…,n, in order to minimize the sum cut of the generated solution. The SumCut problem is really important in archeology (in seriation tasks) and in genetics, helping in the Human Genome Project. This problem is equivalent to the Profile problem, because a solution for SumCut is reversal solution for Profile problem. Experimental results show that the GRASP and Path Relinking methods presented outperform in terms of average percentage deviation the results from the State of the Art using shorter CPU time.

Type
Conference paper
Publication
International Journal of Combinatorial Optimization Problems and Informatics
Jesús Sánchez-Oro
Jesús Sánchez-Oro
Associate Professor

Associate Professor at the Computer Science Department, being one of the senior researchers of the Group for Research on Algorithms For Optimization GRAFO.

Abraham Duarte
Abraham Duarte
Full Professor

Abraham Duarte is Full Professor in the Computer Science Department at the Rey Juan Carlos University (Madrid, Spain). He has done extensive research in the interface between computer science, artificial intelligence, and operations research to develop solution methods based on Computational Intelligence (metaheuristics) for practical problems in operations-management areas such as logistics and supply chains, telecommunications, decision-making under uncertainty and optimization of simulated systems.