This paper presents a survey of novel technologies for uncovering implicit knowledge through the analysis of user-contributed content in Web2.0 applications. The special features of emergent semantics are herein described, along with the various dimensions that the techniques should be able to handle. Consequently a series of application domains is given where the extracted information can be consumed. The relevant techniques are reviewed and categorised according to their capability for scaling, multi-modal analysis, social networks analysis, semantic representation, real-time and spatio-temporal processing. A showcase of such an emergent semantics extraction application, namely ClustTour, is also presented, and open issues and future challenges in this new field are discussed.