An Interaction-Centric Dataset for Learning Automation Rules in Smart Homes

An Interaction-Centric Dataset for Learning Automation Rules in Smart Homes Kai Frederic Engelmann1, Patrick Holthaus1, Britta Wrede1, Sebastian Wrede1,2 1Cluster of Excellence Cognitive Interaction Technology (CITEC), Bielefeld University, Germany 2Research Institute for Cognition and Robotics (CoR-Lab), Bielefeld University, Germany kengelma/pholthau/bwrede/swrede@techfak.uni-bielefeld.de Abstract The term smart home refers to a living environment that by its connected sensors and actuators is capable of providing intelligent and contextualised support to its user. This may result in automated behaviors that blends into the user’s daily life. However, currently most smart homes do not provide such intelligent support. A first step towards such intelligent capabilities lies in learning automation rules by observing the user’s behavior. We present a new type of corpus for learning such rules from user behavior as observed from the events in a smart homes sensor and actuator network. The data contains information about intended tasks by the users and synchronized events from this network. It is derived from interactions of 59 users with the smart home in order to solve five tasks. The corpus contains recordings of more than 40 different types of data streams and has been segmented and pre-processed to increase signal quality. Overall, the data shows a high noise level on specific data types that can be filtered out by a simple smoothing approach. The resulting data provides insights into event patterns resulting from task specific user behavior and thus constitutes a basis for machine learning approaches to learn automation rules. Keywords: Behavioral Corpus, Smart Homes, Internet of Things, Ambient Intelligence 1. Introduction In a recent literature review, Mennicken et al. (2014) ex- amine the current scientific attempts of realizing the con- cept of a smart home. They name the collaboration of en- vironment and user – or shared autonomy – as one of the promising challenges in the field. A system that suggests automatic behaviors for the user is therefore suggested as a possible solution to accomplish satisfying home automa- tion routines. Consequently, a smart environment has to be able to offer users a set of functionalities based on their behavior. Ur et al. (2014) thereby revealed in a study that the function- alities can be enabled on the basis of simple trigger-action rules in more than 60% of the cases. In order to compute these kind of rules, many existing approaches apply tech- niques that learn user behavior. Aztiria et al. (2008), for ex- ample, use machine learning techniques are combined with a speech recognition system that allows for verbal program- ming of the environment. Such a system however still needs manual confirmations or adjustments to control devices like roll-shutters or light systems. Only a few approaches, like MavHome (Cook et al., 2003), are able to skip program- ming routines and learn the rules unsupervised instead. In an intelligent room, residents are observed and the system tries to predict user behavior. We also claim that it is feasible to learn a certain set of rules, e.g. lowering temperatures if the window is open, solely from the observation …

Content truncated to 500 words