Debriefing

Here you can read the debrief after the first meeting with the persons responsible for the project.

Client

• Lectoraat Responsible IT and the HvA Responsible AI Lab
• Product Owner: Yuri Westplat, [email protected]
• Researchers: Marcio Fuckner, Pascal Wiggers

Background information

The Responsible IT research group at the Hogeschool van Amsterdam conducts academic and applied research on technologies behind fake news detection, filter bubbles and recommendation algorithms. Through our collaborations with partners in the Dutch media and design industry and the City of Amsterdam, we recognize the urgent need for an easy-to-use tool like The Bubble Machine to understand the mechanisms of disinformation.

What is bubble machine?

Bubble machine is an online simulator, which gives insight in how persons respond to published articles and how it relates to their filter bubble. The tool is developed to give journalist more insight in how specific bubbles are formed based on different factors in the form of personal preference, social network (friends and colleagues) and technology (algorithm based recommendations).

The assignment

For this project, the goal was to make a new frontend interface for the existing application that is more user-friendly. We're working together with other students from the UX minor. Together we're also responsible for making different and more understandable graphs for the end user.

Cause of the project

At this moment, the simulation is difficult to use, and the product owner would like to bring the application to a wider public.

Goal

The goal is to get familiar with the application and the goal it wants to reach and make a usable front-end experience so it at least a lot easier to use than the current iteration. We want to reach that goal by testing multiple strategies to see what works best by rapid prototyping.

The end user

The end user of the product are journalists that are using the created modal for their own research and to learn more about the spread of information and the creation of bubbles.

Conditions

• Integrating the web application with the Web API, which includes:
• Simulation parameters
• Real-time analysis of population using a graph
• Real-time analysis of metrics (presence of homogeneous networks, diversity metrics)
• Integrated Tests
• User Acceptance Tests
• Deployment of the Web API to the production environment