Collaborative knowledge management and data discovery: The challenges and opportunities ahead

Recent advances in miniaturization of computing devices, communications and digital storage technologies, along with the novel data-acquisition discoveries, have enabled gathering and storing of unprecedented amounts of heterogeneous data [1]. This, in turn, provided foundations for large-scale knowledge discovery and information fusion undertakings addressing issues ranging from serviceoriented distributed data mining in distributed infrastructure-based environments [2], to mining in P2P and sensor networks [3]. To optimize the utilization of the available devices, from their deployment up to the continuous control and collaborative gathering of important data, diverse issues need to be addressed in a collaborative manner, spanning through networking, mobility control, information fusion, signal processing, along with indexing!retrieval, query processing and data mining. One of the main categories of problems stems from the heterogeneity of the devices and the semantics of the different applications that may pose concurrent pending requests. As an example, indexing structures in wireless sensor networks are expected to emerge from the interactions of sensors and be adaptive to topology changes due to sensors being online/offline or by sensor mobility [4,5]. However, detection of particular event and its matching with a rule in the knowledge-base may require a completely "unnatural" re-organization of the structures used and the operating regime of the devices employed.