Abstract

BERNABEL., Maria Beatriz; RODRIGUEZF., Marco Antonio; GONZALEZV., Rogelio  and  ESTRADAA., Martín. A bioinspired proposal based onneighborhoods for partitioning. Rev. Mat [online]. 2016, vol.23, n.1, pp.221-239. ISSN 1409-2433.

One of the main sources of inspiration to propose new computational paradigms has been the observation of nature. Diverse artificial intelligence techniques have been created in this way.

One of the efforts that has caused great impact is imitating the way some living beings survive, and in particular, the study of the study of brain function is useful to propose analogous schemes and solve some problems. In this point, the bioinspired systems have been originated as a set of models based on the behavior of certain biological systems, which can be seen in areas such as data mining and operations research where data clustering stands out. From the need of solving clustering problems, we have proposed a bioinspired neighborhood search partitioning algorithm. This algorithm, under a bioinspired connotation, has been proposed after observing some of the characteristics in common between clustering and human behavior, where said characteristics can be modeled. Given the high complexity of data clustering, we have incorporated variable neighborhood search (VNS) into the bioinspired clustering algorithm. We chose this metaheuristic because of the similarity that exists between VNS and the way that living beings get organized to solve conflict situations.

Keywords : geographical clustering; bioinspired; living beings; neighborhood search.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )