Over the last years, ubiquitous computing has started to cre ate a new world of small, heterogeneous, and distributed devices that have the ability to sense, to co mmunicate and interact in ad hoc or sensor networks and peer2peer systems. These large scale distribu ted systems have in many cases to interact in real-time with their users. Knowledge Discovery in ubiquit o s environments (KDubiq) is an emerging area of research at the intersection of the two major challen ges of highly distributed and mobile systems and advanced knowledge discovery systems. It aims to provid e a unifying framework for systematically investigating the mutual dependencies of otherwise quite u nr lated technologies employed in building next-generation intelligent systems: machine learning, d ata mining, sensor networks, grids, P2P, data stream mining, activity recognition, Web 2.0, privacy, use r modeling and others. In a fully ubiquitous setting, the learning typically takes place in situ, inside th small devices. Its characteristics are quite different from the current mainstream data mining and machi ne learning. Instead of offline-learning in a batch setting, sequential learning, anytime learning, real-time learning, online learning etc. under real-time constraints from ubiquitous and distributed dat a is needed. Instead of learning from stationary distributions, concept drift is the rule rather than the exc eption. Instead of large stand-alone workstations, learning takes place in unreliable, highly resource constr ained environments in terms of battery power and bandwidth. The goal of this special issue is to promote an interdiscipli nary forum for researchers who deal with sequential learning, anytime learning, real-time learnin g, online learning, etc. from ubiquitous and distributed data. Distributed Learning from Data Streams i s a recent and increasing research area with challenging applications and contributions from fields lik e Data Bases, Data Mining, Machine Learning, and Statistics. The selected papers cover a large spectrum i n the research of Ubiquitous Knowledge Discovery that goes from frequent pattern mining algorithm s, distributed clustering, outlier detection to multi-relational learning The common concept in all the p apers is that learning occurs while data continuously flows eventually produced from distributed so urces. The first paper, Gama and Pereira presents a new distributed c l stering algorithm which reduces both the dimensionality and the communication burden between se nsors. Bifet and Gavald à propose new algorithms for adaptively mining closed rooted trees, both labeled and unlabeled, from data streams that change over time. Closed patterns are powerful representat ives of frequent patterns, since they eliminate