Uncertainty handling in mobile community information systems

Current mobile community information systems handle vast amounts of multimedia, various operations on multimedia processing, and diverse user communities using different mobile devices. Imprecise or even false GPS information, users’ false semantic description about a multimedia artifact, or different interpretations of multimedia content in different user communities lead to data uncertainty. Some research work has been conducted in advanced uncertainty databases with probabilistic theories, data lineage, and fuzzy logic. However, these approaches deal with “conventional” data uncertainty problems at the database level. Multimedia content as well as user and community factors are not paid much attention, which is important for data management in prevalent web and mobile community information systems. Hence, this dissertation deals with data uncertainty problems addressed with the emerging and advanced development of Web 2.0 and mobile technologies. In my research I identify the new uncertainty problem perspectives in context, multimedia semantics, and community in mobile community information systems, namely uncertainty 2.0. To deal with this uncertainty 2.0 problem, I have developed a model with a combination of the aspects of context management, semantics management, and community of practice realization. Collaborative tagging for multimedia content and community-based storytelling are the key approaches to handling uncertainty problems and engineering mobile community information systems. This model is established with two foci. First, interactions and conversions between multimedia semantics and multimedia context are crucial for mobile and Web data management in mobile community information systems. Second, development of the concept community of practice has great potentials to reduce uncertainty 2.0 problems. The practices taken by communities consist of multimedia tagging and multimedia storytelling. Amateurs are able to develop their knowledge and experiences in communities of practice. Uncertainty is reduced via this cultivation of expertise with the help of community of practice. A set of mobile community information system applications have been realized in the domains of cultural heritage management and technology enhanced learning, which validates the models and approaches to uncertainty 2.0 handling.