Constructing Adjacency Arrays from Incidence Arrays

Abstract

Graph construction, a fundamental operation in a data processing pipeline, is typically done by multiplying the incidence array representations of a graph, $\mathbf{E}_\mathrm{in}$ and $\mathbf{E}_\mathrm{out}$, to produce an adjacency array of the graph, $\mathbf{A}$, that can be processed with a variety of algorithms. This paper provides the mathematical criteria to determine if the product $\mathbf{A} = \mathbf{E}^{\sf T}_\mathrm{out}\mathbf{E}_\mathrm{in}$ will have the required structure of the adjacency array of the graph. The values in the resulting adjacency array are determined by the corresponding addition $\oplus$ and multiplication $\otimes$ operations used to perform the array multiplication. Illustrations of the various results possible from different $\oplus$ and $\otimes$ operations are provided using a small collection of popular music metadata.

8 pages, 5 figures, accepted to IEEE IPDPS 2017 Workshop on Graph Algorithm Building Blocks