A classic example of a node classification task is to identify bots from a given large social network of millions of users. Examples of node classification beyond social networks include classifying the function of proteins in the interactome and classifying the topic of documents based on hyperlink or citation graphs.

University of Michigan - Ann Arbor

Node classification is a machine learning task on graph data where the goal is to predict the label, which could be a type, category, or attribute, associated with all the nodes, when only given the true labels on a training set of nodes.

Node Classification

Hamilton, W.L. Graph Representation Learning. *Synthesis Lectures on Artificial Intelligence and Machine Learning*. 2020, 14(3), 1–159. https://doi.org/10.2200/S01045ED1V01Y202009AIM046

Graph Representation Learning by William Hamilton

Examples of Node Classification Tasks

The main difference between node classification tasks and traditional supervised classification tasks is that the nodes in a graph are not independent and identically distributed, which is usually assumed in the former case. Instead, we model an interconnected set of nodes, and leverage concepts such as homophily, structural equivalence, and heterophily. Node classification tasks are thus often referred to as semi-supervised.

Learn Before

Related