While knowledge management (KM) is becoming an established discipline with many applications and techniques, its adoption in health care has been challenging. Though, the health care sector relies heavily on knowledge and evidence based medicine is expected to be implemented in daily health care activities; besides, delivery of care replies on cooperation of several partners that need to exchange their knowledge in order to provide quality of care. In public health decision is mainly based on data and a shift is needed towards evidence based decision making. It is obvious that health care can profit from many advantages that KM can provide. Nevertheless, several challenges are ahead, some are proper to KM and other particular to the health care field. This chapter will overview KM, its methods and techniques, and provide and insight into health care current challenges and needs, discuss applications of KM in health care and provide some future perspectives for KM in health care.

Online social networks have been adopted by a large part of the population, and have become in few years essential communication means and a source of information for journalists. Nevertheless, these networks have some drawbacks that make people reluctant to use them, such as the impossibility to claim for ownership of data and to avoid commercial analysis of them, or the absence of collaborative tools to produce multimedia contents with a real editorial value.In this paper, we present a new kind of social networks, namely spontaneous and ephemeral social networks (SESNs). SESNs allow people to collaborate spontaneously in the production of multimedia documents so as to cover cultural and sport events.

Computing pairwise word semantic similarity is widely used and serves as a building block in many tasks in NLP. In this paper, we explore the embedding of the shortest-path metrics from a knowledge base (Wordnet) into the Hamming hypercube, in order to enhance the computation performance. We show that, although an isometric embedding is untractable, it is possible to achieve good non-isometric embeddings. We report a speedup of three orders of magnitude for the task of computing Leacock and Chodorow (LCH) similarity while keeping strong correlations (r = .819, ρ = .826).

In this paper, we introduce a state-of-the-art pseudo-labeling technique for few-shot intent detection. We devise a folding/unfolding hierarchical clustering algorithm which assigns weighted pseudo-labels to unlabeled user utterances. We show that our two-step method yields significant improvement over existing solutions. This performance is achieved on multiple intent detection datasets, even in more challenging situations where the number of classes is large or when the dataset is highly imbalanced. Moreover, we confirm these results on the more general text classification task. We also demonstrate that our approach nicely complements existing solutions, thereby providing an even stronger state-of-the-art ensemble method. We make our code publicly available 1 for future research.

No abstract