Using Temporal Relations in Smart Environment Data for Activity Prediction

The goal of our research is to develop a general framework for representing and reasoning about temporal relations between events in smart home datasets, and to determine the benefits of this reasoning for event prediction. In earlier work, we performed prediction based solely on the sequence of observed activities. In this work, we supplement evidence for a particular action using the temporal relation information. We compare the predictive accuracy with and without temporal information, and illustrate the benefit of temporal relationships for prediction of smart home events and activities.