Access the full text.
Sign up today, get DeepDyve free for 14 days.
R Brits, AP Engelbrecht, F Bergh (2007)
Locating multiple optima using particle swarm optimizationAppl Math Comput, 189
DJ Watts, SH Strogatz (1998)
Collective dynamics of “small-world” networksNature, 393
J Scott (2009)
Social network analysis
SA Hamdan (2008)
Hybrid particle swarm optimiser using multi-neighborhood topologiesINFOCOMP J Comput Sci, 7
R Poli, J Kennedy, T Blackwell (2007)
Particle swarm optimization—an overviewSwarm Intell, 1
J Kennedy, R Mendes (2006)
Neighborhood topologies in fully informed and best-of-neighborhood particle swarmsIEEE Trans Syst Man Cybern Part C Appl Rev, 36
WV Thomas (1996)
Social network thresholds in the diffusion of innovationsSoc Netw, 18
A Bavelas (1950)
Communication patterns in task-oriented groupsJ Acoust Soc Am, 22
T Ray, KM Liew (2003)
Society and civilitation: an optimization algorithm based on the simulation of social behaviorIEEE Trans Evol Comput, 7
M Birattari (2009)
Tuning metaheuristics: a machine learning perspective
RC Eberhart, P Simpson, R Dobbins (1996)
Computational intelligence PC tools
X Li (2010)
Niching without niching parameters: particle swarm optimization using a ring topologyIEEE Trans Evol Comput, 14
The method of musical composition (MMC) is a metaheuristic based on sociocultural creativity systems. Within the MMC, models of social influence and social learning are used and integrated in a social network, which is composed of a set of individuals with links between them and involves a set of interaction rules. In this paper, a comparative study on the performance of the MMC with different network structures is proposed. Sixteen benchmark nonlinear optimization problems are solved, taking into account nine social topologies, which are: (a) linear, (b) tree, (c) star, (d) ring, (e) platoons, (f) von Neumann, (g) full connection, (h) random and (i) small world. In addition, the update of each topology structure was tested according to four different strategies: one static, two dynamic and one self-adaptive states. An exhaustive statistical analysis of the obtained numerical results indicates that the social dynamics has no significant impact on the MMC’s behavior. However, the topology structures can be classified into groups that consistently influence the performance level of the MMC. More precisely, a structure characterized by a low value of its mean number of neighbors and a rather fast information transfer process (star topology) performs in a radically opposite way as structures where each agent has many neighbors (random and complete topologies). These observations allow to provide some guidelines for the selection of a network topology used within a social algorithm.
Artificial Intelligence Review – Springer Journals
Published: Jan 23, 2016
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.