COLLABORATIVE RULE-BASED PROACTIVE SYSTEMS: MODEL, INFORMATION SHARING STRATEGY AND CASE STUDIES

The Proactive Computing paradigm provides us with a new way to make the multitude of computing systems, devices and sensors spread through our modern environment, work for/pro the human beings and be active on our behalf. In this paradigm, users are put on top of the interactive loop and the underlying IT systems are automated for performing even the most
complex tasks in a more autonomous way. This dissertation focuses on providing further means, at both theoretical and applied levels, to design and implement Proactive Systems. It is shown
how smart mobile, wearable and/or server applications can be developed with the proposed Rule-Based Middleware Model for computing pro-actively and for operating on multiple platforms. In order to represent and to reason about the information that the proactive system needs to know about its environment where it performs its computations, a new technique called Proactive Scenario is proposed. As an extension of its scope and properties, and for achieving global reasoning over inter-connected proactive systems, a new collaborative technique called Global Proactive Scenario is then proposed. Furthermore, to show their potential, three real world case studies of (collaborative) proactive systems have been explored for validating the proposed
development methodology and its related technological framework in various domains like e-Learning, e-Business and e-Health. Results from these experiments con rm that software applications designed along the lines of the proposed rule-based proactive system model
together with the concepts of local and global proactive scenarios, are capable of actively searching for the information they need, of automating tasks and procedures that do not require the user's input, of detecting various changes in their context and of taking measures to adapt to it for addressing the needs of the people which use these systems, and of performing collaboration and global reasoning over multiple proactive engines spread across different networks.