Paper accepted to LoG 2024
Our paper has been accepted for an oral presentation at the Learning on Graphs Conference (LoG 2024).
Homophily is a graph property describing the tendency of edges to connect similar nodes. There are several measures used for assessing homophily but all are known to have certain drawbacks: in particular, they cannot be reliably used for comparing datasets with varying numbers of classes and class size balance. To show this, previous works on graph homophily suggested several properties desirable for a good homophily measure, also noting that no existing homophily measure has all these properties. Our paper introduces a new homophily measure — unbiased homophily — that has all the desirable properties and thus can be reliably used across datasets with different label distributions. We show the advantages of unbiased homophily over other existing measures both theoretically and via empirical examples.