Labeled networks and network learning

Predictive Analytics using Networked Data in R

María Óskarsdóttir, Ph.D.

Post-doctoral researcher

customers

    id churn
1    1     0
2  393     0
3 2573     0
4 4430     0
5  926     1
6 1574     1

edgeList

  from    to
1    1   393
2    1  2573
3    1  4430
4  393   926
5  393  1574

Churn

The Relational Neighbor Classifier

Neighbors of Cecelia
- A,B,D,G
Neighbors of Cecelia that prefer R
- A, B, D (75%)
Neighbors of Cecelia that prefer Python
- G (25%)
Cecelia has a higher probability to prefer R

The Relational Neighbor Classifier

rNeighbors <- c(4,3,3,5,3,2,3,0,1,0)
pNeighbors <- c(0,0,1,1,0,2,2,3,3,2)

rRelationalNeighbor <- rNeighbors / (rNeighbors + pNeighbors)
rRelationalNeighbor

1.00 1.00 0.75 0.86 1.00 0.50 0.60 0.00 0.00 0.00

Let's practice!

Predictive Analytics using Networked Data in R