exit

Carlos M. Fernandes
LARSyS - University of Lisbon
Portugal



Publié le 18 octobre 2017   DOI : 10.21494/ISTE.OP.2017.0179

Résumé

Abstract

Mots-clés

Keywords

This paper provides technical and conceptual information on a swarm art project called Sleep Paintings. The project consists of a series of colour images generated by an ant-based clustering algorithm operating with sleep data sets extracted from electroencephalogram (EEG) signals. The algorithm was originally proposed for clustering and classification and is described by a simple set of equations that model the local behavior of artificial ants (data samples) in a way that, when travelling on a heterogeneous 2-dimensional lattice of vectors, they form clusters of similar data samples, while changing the vectors of the environmental lattice. The 2-dimensional colored abstract sketches of human sleep described in the paper result from the visualization of the lattice.

This paper provides technical and conceptual information on a swarm art project called Sleep Paintings. The project consists of a series of colour images generated by an ant-based clustering algorithm operating with sleep data sets extracted from electroencephalogram (EEG) signals. The algorithm was originally proposed for clustering and classification and is described by a simple set of equations that model the local behavior of artificial ants (data samples) in a way that, when travelling on a heterogeneous 2-dimensional lattice of vectors, they form clusters of similar data samples, while changing the vectors of the environmental lattice. The 2-dimensional colored abstract sketches of human sleep described in the paper result from the visualization of the lattice.

swarm art generative artificial ants ant algorithms data analysis sleep

swarm art generative artificial ants ant algorithms data analysis sleep