Context-Aware Mobile Computing:
Learning Context-Dependent Personal Preferences from a Wearable Sensor Array
Aufgabensteller: Prof. Bernd Brügge, Ph.D.?
Prof. Daniel P. Siewiorek, Ph.D. and MartinWagner
Context-aware mobile computing describes the situation where a wearable / mobile computer is aware of its user's state and surroundings and modifies its behavior based on this information.
In this thesis, a method was developed and based thereupon, a wearable system was designed, implemented and evaluated which can learn context-dependent personal preferences by identifying individual user states and observing how the user interacts with the system in these states.
This learning occurs online and does not require external supervision.
The system relies on techniques from artificial intelligence, machine learning and statistical analysis.
A case study investigates the appropriateness of this approach for SenSay
, a context-aware mobile phone implementation and examines why existing methods, which rely on predefined rule-sets, are not suitable for this particular application.
The results indicate the power of the method to automatically determine a model for context-sensitive response for the investigated application while only requiring data from an array of comfortable wearable sensor devices.