Real-world is inherently uncertain. Uncertainty indicates the lack of confidence in an event or decision. It arises from different sources and in various forms. In the context of activity recognition, uncertainty may be due to sensor errors, communication failures and variability in human activities. Reasoning under uncertainty is a process of deducing new knowledge based on available but incomplete information. The purposes of reasoning under uncertainty can be divided into two: improving the quality of context information and inferring new kind of information [1]. Quality of context information quantifies the quality characteristics of the context information using quality attributes [2] such as precision, probability of correctness, trust-worthiness, resolution, up-to-dateness. Usually it takes the form of multi-sensor fusion where data from difference sensors are used to increase confidence.
[1] C. Bettini, O. Brdiczka, K. Henricksen, J. Indulska, D. Nicklas, A. Ranganathan, and D. Riboni, “A survey of context modelling and reasoning techniques,” Pervasive and Mobile Computing, vol. 6, no. 2, pp. 161–180, Apr. 2010.
[2] R. Neisse, M. Wegdam, and M. van Sinderen, “Trustworthiness and Quality of Context Information,” in Young Computer Scientists, 2008. ICYCS 2008. The 9th International Conference for, 2008, pp. 1925–1931.