With the rapid evolution of mobile devices, the concept of context aware applications has gained a remarkable popularity in recent years. Smartphones and tablets are equipped with a variety of sensors including accelerometers, gyroscopes, and GPS, pressure gauges, light and GPS sensors. Additionally, the devices become computationally powerful which allows real-time processing of data gathered by their sensors. Universal network access via WiFi hot-spots and GSM network makes mobile devices perfect platforms for ubiquitous computing. Most of existing frameworks for context-aware systems, are usually dedicated to static, centralized, clientserver architectures. However, mobile platforms require from the context modeling language and inference engine to be simple and lightweight. The model should also be powerful enough to allow not only solving simple context identification tasks but more complex reasoning. The original contribution of the paper is a proposal of a new rule-based context reasoning platform tailored to the needs of such intelligent distributed mobile computing devices. It contains a proposal of a learning middleware supporting context acquisition. The platform design is based on a critical review and evaluation of existing solutions given in this paper. A preliminary evaluation of the platform is given along with use cases including a social system supporting crime detection and investigation.