Problem Description

We are interested in intelligent decision support systems that we often call “assistance systems”. There could be avatars in VR that help you when you have to make a decisions. In our setup, these are often shopping decisions but it could also be anything else. We like to use the EMOTIBIT device in order to predict cognitive load. Emotibiti is a wearable sensor module for capturing high-quality emotional, physiological, and movement data. This project offers you the opportunity to conduct an experiment, collect data and train a machine learning classifier on the collected data. You first inform yourself about related work regarding the topics “experiment design”, “classification using machine learning”, and “biosignals”.

Goal of the Thesis

Collect data with the EMOTIBIT device and train a (multiclass) machine learning classifier. The target variable for the classification should be cognitive load. you can make use of three cognitive load tasks that we have already programmed in VR and record the user data with EMOTIBIT while conducting these three tasks.If you do not want to dive in VR then you can also re-programm these tasks in a web interface and make the recording on an ordinary web interface. Use the Lab Syncing Layer (LSL) in combination with the BrainFlow library.

Requirements

• Basic programming skills
• Maybe willingness to learn some C++ basics to dig into the boost library
• Machine Learning Basics (classifier can be XGBoost, but a bonus could be a more advanced classifier that uses time series instead of tabular data)

Sources

• Montgomery, S. M., Nair, N., Chen, P., & Dikker, S. (2024). Validating EmotiBit, an open-source multi-modal sensor for capturing research-grade physiological signals from anywhere on the body. Measurement: Sensors, 32, 101075.
• Schultz, T., & Maedche, A. (2023). Biosignals meet adaptive systems. SN Applied Sciences, 5(9), 234.
• Weiß, T., & Pfeiffer, J. (2024). Consumer decisions in virtual commerce: Predict good help-timing based on cognitive load. Journal of Neuroscience, Psychology, and Economics, 17(2), 119.
• Li, H. (2023). Machine Learning Methods. Springer Singapore. 
• https://www.emotibit.com/