Partitioning sparse matrices with eigenvectors of graphs
SIAM Journal on Matrix Analysis and Applications
EDA-USL: unsupervised clustering algorithm based on estimation of distribution algorithm
International Journal of Wireless and Mobile Computing
International Journal of Wireless and Mobile Computing
An evolvable cellular automata based data encryption algorithm
International Journal of Wireless and Mobile Computing
Hi-index | 0.00 |
To overcome the drawbacks of traditional complex community network division algorithms, such as low accuracy, high time complexity and easily being trapped into local optimum, a novel complex community network division algorithm is proposed by using Multi-Gene Families MGF. The proposed approach first encodes the node ID and the community type into two different MGFs, respectively, and then encodes the relationship of the two MGFs into the chromosome through a mapping function. Moreover, in order to prevent premature and speed up convergence, the elite migration strategy is utilised throughout the evolution process, such as gene selection, chromosome crossover, chromosome inversion, restricted permutation. The experiments and analyses show that our approach is better than the traditional evolutionary algorithm in terms of efficiency and accuracy.