Logo des Repositoriums

Auflistung nach Autor:in "Knierim, Michael"

1 - 3 von 3
  • Konferenzbeitrag
    Classification of Music Preferences Using EEG Data in Machine Learning Models
    (Mensch und Computer 2024 - Workshopband, 2024) Vedder, Helen; Stano, Fabio; Knierim, Michael
    In this paper, we investigate how EEG data can be used to predict individual music preferences. Our study relies on machine learning and specially developed models such as EEGNet to analyze participants' brain activity while listening to music. Participants listened to music excerpts, rated them, and their EEG data were recorded. We extracted relevant features from the EEG data and used convolutional neural networks (CNNs) to classify music preferences. Our results show that our models are able to predict music preferences with an accuracy of up to 69%. This confirms the potential of EEG in personalized music recommendation and demonstrates the feasibility of integrating EEG into wearable devices to improve the user experience.
  • Zeitschriftenartikel
    Hybrid Adaptive Systems
    (Business & Information Systems Engineering: Vol. 66, No. 2, 2024) Benke, Ivo; Knierim, Michael; Adam, Marc; Beigl, Michael; Dorner, Verena; Ebner-Priemer, Ulrich; Herrmann, Manfred; Klarmann, Martin; Maedche, Alexander; Nafziger, Julia; Nieken, Petra; Pfeiffer, Jella; Puppe, Clemens; Putze, Felix; Scheibehenne, Benjamin; Schultz, Tanja; Weinhardt, Christof
  • Konferenzbeitrag
    OpenEarable Suite: Open-Source Hardware to Sense 30+ Phenomena on the Ears
    (Mensch und Computer 2024 - Workshopband, 2024) Röddiger, Tobias; Knierim, Michael; Lepold, Philipp; King, Tobias; Beigl, Michael
    In this demo, we showcase the OpenEarable Suite, a comprehensive collection of ear-worn devices designed to sense and analyze over 30 different phenomena. The collection includes three distinct devices: OpenEarable 2.0, OpenEarable ExG, and OpenEarable ExG Headphones. "OpenEarable 2.0" integrates advanced sensors, such as ultrasound-capable microphones, a 9-axis inertial measurement unit, a pulse oximeter, an optical temperature sensor, and an ear canal pressure sensor, enabling extensive health monitoring, activity tracking, and human-computer interaction. "OpenEarable ExG" is an open-source platform focused on measuring biopotentials like EEG, ECG, and EMG, using up to four sensing channels, and validated for detecting eye movements, brain activity, and muscle contractions. "OpenEarable ExG Headphones" combine electrophysiological sensing with high-quality audio, utilizing OpenBCI biosignal amplification and a 3D-printed over-ear design for reliable EEG, EOG, ECG, and EMG measurements. The OpenEarable Suite aims to democratize earable research by providing accessible, open-source tools in different form factors that follow best practices in hardware and software development, facilitating diverse applications across various domains from medical to HCI.