Resumo

REYNA-HERNANDEZ, Randy  e  ROSETE, Alejandro. Experimental determination of the influence of the reformulation of the RAP problem on its computational efficiency. Rev cuba cienc informat [online]. 2022, vol.16, n.1, pp. 62-76.  Epub 01-Mar-2022. ISSN 2227-1899.

The Rank Aggregation Problem (RAP) search for a ranking that summarizes a set of them. Currently, the reasoning based on rankings has gained great attention due to the large number of applications to solve decision-making problems and more recently, in computer science, statistics, linear algebra and optimization, computational biology among many others. A particular case of RAP is the Kemeny Ranking Problem (KRP), where all entry rankings are a permutation. The KRP is NP-hard, however, there is a formulation where, through the method of Integer Linear Programming (PLE), the problem can be solved. As the efficiency of the PLE method is given by the number of variables and restrictions, a reformulation of the problem was obtained that allows both dimensions to be reduced. In the present work it is demonstrated how the reformulation of the RAP allows to solve with the same software major instances and to solve them in less time.

Palavras-chave : Ranking Aggregation; Integer Linear Programming; KNIME Plugins; Kemeny Ranking Problem.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )