Problem Description

Learn about the problem of node classification in a contact tracing network.

Contact tracing

Contact tracing was a popular term during the COVID-19 pandemic. It's a process of identifying people who were in close contact with infected individuals. The aim is to identify the individuals quickly and isolate them to control the spread.

There are many ways to do contact tracing, which involves using different techniques. Most contact tracing methods are crowdsourced and operate using voluntary participants. In this case study, we'll learn how to formulate a node classification task on a contact tracing network using graph neural networks.

Let's construct a simulated contact tracing graph for the purpose of node classification. The graph should contain the following details:

  • Name and details of the person

  • Test results

  • Symptoms

  • Mobility information

  • Vaccination information

  • Details of people in close contact

These are some of the details necessary to build a contact tracing network.

The dataset

We can generate a contact tracing graph using the Watts–Strogatz model. Let's take a look at the code below:

Get hands-on with 1200+ tech skills courses.