Related Projects

In openSMILE ( a feature extraction tool was developed to extract audiovisual features related to affective behaviours. This tool was successfully used as a baseline in two series of international challenges for paralinguistic and affect analysis from speech (INTERSPEECH ComParE 2009-2014) and audiovisual data (ICMI AVEC 2010-2014). Models of affective behaviors can easily be added, which facilitates the integration of new languages. openSMILE will be used for automatic analysis MixedEmotions 10 of affective behaviours from audiovisual data.

logo_grandeEUROSENTIMENT ( provides a shared language pool for sentiment and emotion analysis from different SMEs, following a linked data approach. In addition, EUROSENTIMENT provides a semantic pipeline for entity-level sentiment lexicon generation, following a linked data approach for lexical resources (vocabulary lemon), Sentiment expressions (vocabulary Marl) and emotion analysis (vocabulary Onyx compatible with EmotionML). In addition, EUROSENTIMENT defines NIF REST services so that it can be combined with other NLP services. The project is lead by Paradigma with participation of NUIG, SindiceTech and UPM.

JUNIPER ( is a FP7 big-data project (BUT participates) that provides a platform supporting a range of high-performance application domains that seek real-time processing of streaming data or real-time access to stored data. MixedEmotions will particularly benefit from the combinations linking standard map-reduce frameworks (such as Hadoop) with stream-based processing (such as Apache Storm).

Interactive Knowledge Stack (IKS) ( is a recently finished EU Integrated Project with a focus on providing an open source technology platform for semantically enhanced content management systems and resulted in an open source community and projects e.g., Apache Stanbol (modular software stack and reusable set of components for semantic content management) (, VIE (Vienna IKS Editables: semantic web applications and user interaction development platform) (, where relevant staff at Attensity cooperated in