I have a social network(graph) of group of people in which all nodes considered as individual person and links between them shows, if the communication established between them. Here i would like to identify most centralized node in that network using genetic algorithm. Is that possible?
Normally one would calculate betweenness by choosing pairs of vertices and incrementing a counter for each vertex on a shortest path. A randomized version might use a drunken walker. A random walk could express a genome. Two genomes, if they share a vertex, can produce offspring by a crossover there. In the end, you've got to sum up all occurrences of all vertices over all genomes of the pool.
Regards,
Joachim
https://www.cse.unr.edu/~sushil/pubs/newestPapers/2011/cec/cec2011v3.pdf
https://dl.acm.org/citation.cfm?id=1570068
Article Genetic clustering of social networks using random walks
Article Introducing a procedure for developing a novel centrality me...
Genetic algorithms are good in optimization problems. May be it will better for you to go ahead with some classification method.
- Badges
- Science topic
- Similar topics
- Journal Impact Factor