Authors:
Matthias Honal
1
and
Tanja Schultz
2
Affiliations:
1
Carnegie Mellon University, United States
;
2
Carnegie Mellon University; Karlsruhe University, Germany
Keyword(s):
Human-centered systems, Brain Activity, EEG, Task Demand Identification, Meeting and Lecture Scenario.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Computational Intelligence
;
Data Manipulation
;
Devices
;
Health Engineering and Technology Applications
;
Health Information Systems
;
Human-Computer Interaction
;
Methodologies and Methods
;
Neural Networks
;
Neurocomputing
;
Neurotechnology, Electronics and Informatics
;
Pattern Recognition
;
Physiological Computing Systems
;
Physiological Processes and Bio-Signal Modeling, Non-Linear Dynamics
;
Real-Time Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
;
Theory and Methods
;
Wearable Sensors and Systems
Abstract:
Our society demands ubiquitous mobile devices that offer seamless interaction with everybody, everything, everywhere, at any given time. However, the effectiveness of these devices is limited due to their lack of situational awareness and sense for the users’ needs. To overcome this problem we develop intelligent transparent human-centered systems that sense, analyze, and interpret the user’s needs. We implemented learning approaches that derive the current task demand from the user’s brain activity by measuring the electroencephalogram. Using Support Vector Machines we can discriminate high versus low task demand with an accuracy of 92.2% in session dependent experiments, 87.1% in session independent experiments, and 80.0% in subject independent experiments. To make brain activity measurements less cumbersome, we built a comfortable headband with which we achieve 69% classification accuracy on the same task.