Back to Volume
Paper: Searching for Data: Swarming Agent Method
Volume: 453, Advances in Computational Astrophysics: Methods, Tools, and Outcome
Page: 349
Authors: Caputo, D. P.; Dolan, R.
Abstract: As our ability to produce data grows our ability to examine and find the useful portions of large data sets must grow as well. We present an efficient, agent based search algorithm, based on the behavior of schooling fish in the presence of predators, designed to search and/or map very large data sets. Our algorithm, which belongs to the artificial life family of algorithms, attempts to leverage swarm intelligence against the difficulty of finding valuable data within a sea of data. The agents search the data space based on a small set of simple rules which produces emergent behavior and results in an efficient and flexible algorithm, while at the same time resisting many of the short comings of other artificial life algorithms.
Back to Volume