Advanced search


Luis A. Leiva. Human-Computer Interaction meets Machine Learning. 2015. Telefónica I+D, Barcelona

Our ability to solve challenging research problems usually relies on a mix of human and machine intelligence. Human-computer interaction (HCI) is particularly focused on the interfaces between people and computers, whereas machine learning (ML) provides both theoretical and empirical tools for exploiting machine intelligence. The intersection of these communities yields both new opportunities and challenges for a wealth of research domains. In this talk, I will present different projects developed recently at the PRHLT research center aimed at bringing together HCI and ML. Concretely, I will give four lighting talks about: * Warped K-Means: an algorithm to cluster sequentially-distributed data. * MinGestures: a context-aware gesture recognizer for mixed-initiative text editing interfaces. * MemRep: a sampling method for selecting representative and memorable sentences. * μcaptcha: Human Interaction Proofs (HIPs) tailored to touch-capable devices via math handwriting.