Master Project Topics

While artificial intelligence is often viewed as a competitor to human intelligence, an interesting open challenge is to combine the advantages of both by developing mechanisms for cooperative human-AI decision-making. A systematic study of those mechanisms requires experiments with groups of human experts, whose decision-making process is assisted by machine learning techniques.

Project Goals

The goal of this Master Project is a fully documented prototype of a web-based platform that can be used to perform experiments on group decision-making in mixed teams of human experts and AI agents: 

• implementation of different group decision-making strategies (consensus, majority vote, delphi method)
• support for different interaction mechanisms (text chat, display of other decisions) and topologies (flat, hierarchy, network, subgroups)
• rich HTML5/d3js interface with support for different decision tasks implemented as exchangeable modules
• proof-of-concept decision task: image classification under uncertainty
• possibility to engage in a collaboration addressing medical diagnostics together with researchers at the Universitätsspital Zürich

This project provides you with an opportunity to gain experience with state-of-the-art data analytics and machine learning algorithms as well as web technologies. If you are interested, please contact us via E-mail.