Summary

We started this chapter by explaining our problem and dataset and conducted product categorization using network/graph theory. We introduced the R package, igraph, to create and manipulate graphs. We set out discovering some of the properties of the graph and also a little bit about the graphs and random walks. We used the arules package to find frequent pairs of items and finally applied the graph clustering algorithm to our data to discover product categories. We were also able to identify important vertices performing the role of the core and connectors.