This function is based on the concept of multilayer distance. This concept generalizes single-layer distance to a vector with the distance traveled on each layer (in the "multiplex" case). Therefore, non-dominated path lengths are returned instead of shortest path length, where one path length dominates another if it is not longer on all layers, and shorter on at least one. A non-dominated path length is also known as a Pareto distance. Finding all multilayer distances can be very time-consuming for large networks.
distance_ml(n, from, to=character(0), method="multiplex")A data frame with one row for each non-dominated distance, specifying the number of steps in each layer.
A multilayer network.
The actor from which the distance is computed.
The actor(s) to which the distance is computed. If not specified, all actors are considered.
This argument can take values "simple", "multiplex", "full". Only "multiplex" is currently implemented.
Magnani, Matteo, and Rossi, Luca (2013). Pareto Distance for Multi-layer Network Analysis. In Social Computing, Behavioral-Cultural Modeling and Prediction (Vol. 7812, pp. 249-256). Springer Berlin Heidelberg.
multinet.actor_measures, multinet.layer_comparison
net <- ml_aucs()
distance_ml(net,"U54","U3")
Run the code above in your browser using DataLab