Bipartition of small-world networks

• Autor(a) principal: Vieira, José Vítor Correia Rendeiro
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10773/27755

We studied the bipartitioning of small-world networks. We generated small-world networks from a square lattice via a modified Watts- Strogatz algorithm. We compared several partitioning algorithms, such as Monte Carlo with Kawasaki dynamics and Simulated Annealing, Extremal Optimization and Multilevel K-way partitioning. We obtained the critical percolation values of the mean degree and the threshold partition values for several networks in the continuum between a square lattice and a random network. We obtained the exponents of the minimum partition cost as a function of the mean degree in the vicinity of the bipartition threshold, as well as the exponent of finite size scaling. To the best of our knowledge, this is the first work to tackle this issue. We observed that all small-world networks have the same exponents, different from those of the square lattice, regardless of the number of modified edges. The values for the exponents of the random network are in accordance with previous results.