- Full text PDF available (11)
Data Set Used
The authors detail the alignment prediction approach-a time-series-estimation technique applicable to both numeric and nonnumeric data-and compare it to four other prediction approaches to determine context-prediction accuracy in ubiquitous computing environments.
Context awareness is a key feature of modern computing applications, allowing application behaviour to be adapted to context. The power and quality of context-aware applications can significantly be increased by not only considering past and present contexts for adaptation, but by also predicting and reacting to future contexts. This paper deals with… (More)
Context prediction is the task of inferring information about the progression of an observed context time series based on its previous behaviour. Prediction methods can be applied at several abstraction levels in the context processing chain. In a theoretical analysis as well as by means of experiments we show that the nature of the input data, the quality… (More)
The popularization of mobile devices in the last years has led to the development of an increasing number of services. However the technical limitations of mobile devices call for services that require minimal interactions with the user and adapt their behaviors to the user's expectations. Context-awareness has proven to facilitate personalization of… (More)
We propose Foxtrot, a highly flexible architecture for context aware systems that adapts to the versatile requirements of ubiquitous computing environments. To further enhance the gains of the architecture we introduce a context prediction paradigm that incorporates the possibility to increase the accuracy of proactive context aware applications. To utilise… (More)
The ability to predict future contexts significantly expands the possibilities of context-aware computing applications. However , an incorrect prediction may also mislead the application and may result in inappropriate application behaviour. We study influences on the prediction accuracy and propose a novel approach to context prediction in ubiquitous… (More)