Engineering Contextual Knowledge for Autonomic Pervasive Services

Services for mobile and pervasive computing should extensively exploit contextualinformation both to adapt to user needs and to enable autonomic behavior. This raises theproblem of how to represent, organize, aggregate, and make available such data to servicesso as to have it become meaningful and usable knowledge, facilitating the design anddevelopment of autonomic pervasive services, and enabling them to acquire high-degrees ofcontext awareness at limited efforts. In this paper, we identify the key software engineeringchallenges introduced by the need of accessing and exploiting huge amount ofheterogeneous contextual information. Following, we survey the relevant proposals in thearea of context-aware pervasive computing, data mining and granular computing discussingtheir potentials and limitations with regard to their adoption in the development of contextawarepervasive services. On these bases, we propose the W4 model for contextual data andshow how it can represent a simple yet effective model to enable flexible general-purposemanagement of contextual knowledge, to facilitate services in achieving high degrees ofcontext-awareness and, overall, to facilitate the design and development of complexpervasive services. A summarizing discussion and the identification of open researchdirections conclude the paper.