Neighborhood link-based features
Sometimes, the feature values of neighboring nodes have an effect on behavior. In this exercise, you will look at the attribute value of neighboring nodes and compute their average. You will do this for degree, triangles, transitivity, and betweenness.
You need to:
- Multiply the adjacency matrix with the network attribute you want to find the average of, to obtain the overall value in the neighborhood.
- To get the average, divide by the node's degree, given by the vector
degree
which has been pre-loaded. - Finally, convert the result to a vector and assign to
network
as a node attribute.
This exercise is part of the course
Predictive Analytics using Networked Data in R
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Extract the average degree of neighboring nodes
V(network)$averageDegree <-
as.vector(AdjacencyMatrix %*% V(network)$___) / degree