A Probabilistic Model for Node Classification in Directed Graphs

TL;DR

This paper introduces a probabilistic generative model for node classification in directed graphs with attributes and labels, offering interpretability and competitive accuracy, including new benchmarks on a Math Genealogy dataset.

Contribution

The paper presents a novel probabilistic model for directed graph node classification that is interpretable and performs competitively with state-of-the-art methods, also providing benchmarks on a new dataset.

Findings

Model achieves competitive or superior accuracy compared to existing methods.

Demonstrates interpretability of predictions through a probabilistic framework.

Provides new benchmarks on the Math Genealogy dataset.

Abstract

In this work, we present a probabilistic model for directed graphs where nodes have attributes and labels. This model serves as a generative classifier capable of predicting the labels of unseen nodes using either maximum likelihood or maximum a posteriori estimations. The predictions made by this model are highly interpretable, contrasting with some common methods for node classification, such as graph neural networks. We applied the model to two datasets, demonstrating predictive performance that is competitive with, and even superior to, state-of-the-art methods. One of the datasets considered is adapted from the Math Genealogy Project, which has not previously been utilized for this purpose. Consequently, we evaluated several classification algorithms on this dataset to compare the performance of our model and provide benchmarks for this new resource.