Getting a flat dataset
In this exercise, you will turn your network into a dataframe, where the rows are the people in the network and the columns are the network features you computed in the previous chapter. You will also prepare the dataset for the pre-processing.
This exercise is part of the course
Predictive Analytics using Networked Data in R
Exercise instructions
- Extract the dataframe of the customers using the
as_data_frame()
function. Note that you want the node attributes, i.e. vertices. Call the datasetstudentnetworkdata_full
- Inspect the first few rows of the data frame using the
head()
function. - Remove the customers who already churned by conditioning on the
Churn
attribute. Call this dataframestudentnetworkdata_filtered
- Remove the first two columns, called Churn and name, since you don't need them anymore and name the dataframe
studentnetworkdata
.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Extract the dataset
studentnetworkdata_full <- ___(network, what = ___)
# Inspect the dataset
head(___)
# Remove customers who already churned
studentnetworkdata_filtered <- studentnetworkdata_full[-which(studentnetworkdata_full$___ == 1), ]
# Remove useless columns
studentnetworkdata <- ___[, -c(1, 2)]