Search results

This paper aims to apply a contextual case-based reasoning (CBR) to a mobile device. The CBR method was chosen because it does not require training, demands minimal processing resources and easily integrates with the dynamic and uncertain nature of pervasive computing. Based on a mobile user’s location and activity, which can be determined through the device’s inertial sensors and GPS capabilities, it is possible to select and offer appropriate services to this user.

The proposed approach comprises two stages. The first stage uses simple semantic similarity measures to retrieve the case from the case base that best matches the current case. In the second stage, the obtained selection of services is then filtered based on current contextual information.

This two-stage method adds a higher level of relevance to the services proposed to the user; yet, it is easy to implement on a mobile device.

A two-stage CBR using light processing methods and generating context aware services is discussed. Ontological location modeling adds reasoning flexibility and knowledge sharing capabilities.

The prediction of a context, especially of a user’s location, is a fundamental task in the field of pervasive computing. Such predictions open up a new and rich field of proactive adaptation for context-aware applications. This study/paper aims to propose a methodology that predicts a user’s location on the basis of a user’s mobility history.

Contextual information is used to find the points of interest that a user visits frequently and to determine the sequence of these visits with the aid of spatial clustering, temporal segmentation and speed filtering.

The proposed method was tested with a real data set using several supervised classification algorithms, which yielded very interesting results.

The method uses contextual information (current position, day of the week, time and speed) that can be acquired easily and accurately with the help of common sensors such as GPS.