Hi, I'll explain very briefly what work. Iwork with algorithms for semi-supervised learning based on graphs. The process works as follows: I carry a set of data from the UCI Repository and I store the values in an array, then I calculate the similarity between examples using a distance function any pre-set, calculated values are stored in a new array and are used as weights for the connections between the instances (vertices of the graph). The graph has been generated, first apply selection algorithms working and then propagating labels. What I do is: 1) give away a pair of vertices and connect them with weights defined by the similarity between the samples, which was calculated previously; 2) repeating the process until an average degree to be obtained, ranging from 1 to 5 (graphs generated during this step and should not contain loops or multiple edges, so the need to simplify use () function) 3) save the graphs for each average degree obtained I sincerely hope you have been able to explain to you what needs to be done and you can help me.
Thank you!
_______________________________________________ igraph-help mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/igraph-help
