A flexible and configurable middleware for context-aware access to heterogeneous resources

Overwhelming quantity of information on the Internet has complicated the user's experience when trying to find information suitable to their context. Context-Aware Mobile mashUpS (CAMUS) proposes a framework capable of seamless data integration of resources obtained from multiple heterogeneous sources while applying context-based filtering of this data. This thesis proposes a complete design and architecture for the CAMUS system, including a middleware that creates a web API and enables context-awarness following CAMUS concepts. Domain independence and flexibility have been kept in mind while setting the guidelines of this project. Through an iterative process, an implementation of the middleware is developed. The introduction of a cache has proven to be essential to enable the communication between the data integration and context-awarness components, and to improve the query time. Manual and automated tests, in addition to load simulations, have been performed to assess the feasibility and performance of the proposed system. A prototype of the administrative interface to modify the configuration is also illustrated.