Kleenks: Linked Data with Applications in Research and Ambient Assisted Living Andrei-Adnan Ismail 1 , Razvan Dinu 2 Adina Magda Florea 1 , Tiberiu Stratulat 2 , and Jacques Ferber 2 1 University Politehnica of Bucharest, 060042, 313 Splaiul Independentei andrei.ismail@cs.pub.ro http://aimas.cs.pub.ro/people/andrei.ismail 2 University of Montpellier 2, UMR 5506 - CC477, 161 rue Ada, Montpelllier, France dinu@lirmm.fr http://www.lirmm.fr/ ˜ dinu/ Abstract. In the spirit of the Linked Data initiative pioneered by Tim Berners- Lee, we propose a new type of link between entities identified by URIs named kleenk. This new type of link bridges the gap between the classical structured data published as RDF and the semi-structured data formats pushed by social networks such as Facebook and Twitter. Unlike previously published work on RDF and its extensions, our proposal promotes links to a first-class citizenship in the world of a semantic web: a kleenk has a content of itself, can be linked to recursively, and both people and machines can harmoniously collaborate in creating, evaluating and extending them. We discuss the theoretical model of the kleenk and how it can be built on top of existing frameworks such as RDF, and identify the main challenges for adoption of the new model. Finally, we validate our model with two real-world implementations: an online platform for spreading research results between researchers and an ambient intelligence elder tracking scenario where kleenks are used to perform sensor fusion between heterogeneous data sources. Keywords: linked data, ambient intelligence, RDF. 1 Introduction “This is what Linked Data is all about: it’s about people doing their bit to produce a little bit, and it’s all connecting.,, - Tim Berners-Lee, TED 2009. Linked data is a movement trying to expose the world’s data in a structured format and to link it all together in meaningful ways. This concept has been gaining traction as more and more organizations are starting to expose their data in a structured, computer- understandable format, besides the traditional website. Until recent, the habit was this: if an organization owned some data and it wanted to expose it to the public, it created a website allowing users to explore it. However, it soon became obvious that this was not enough; humans were not the only ones interested in working with this data, sometimes even computers or software agents delegated by humans should be able to manipulate E. Mercier-Laurent and D. Boulanger (Eds.): AI4KM 2012, IFIP AICT 422, pp. 53–71, 2014. c  IFIP International Federation for Information Processing 2014