MixedEmotions: A Big Data Platform for Multilingual and Multimodal Emotion Analysis

Monday, September 12, 2016 - 09:00 to 17:00
Campus Augustusplatz

ChairsPaul Buitelaar (Insight-Centre / NUIG)


Emotion analysis is central to tracking customer behavior and satisfaction in the form of email, call center interaction, social media comments, etc. and can be detected through the analysis of facial expression, speech or audio interaction. Making sense of user interaction in different modalities and languages is challenging and has not yet been explored in fullness in an industrial context. In  MixedEmotions we are developing an open source big data platform for semantic-level emotion detection in multilingual and multimodal data streams that will enable SMEs to explore this promising area. The MixedEmotions platform comprises a number of integrated components and  is developed and validated in three proof of concept demonstrators (Social TV, Brand Reputation Management, Call Centre Operations). It will be made available to the community as open source by the end of the project.

The tutorial will guide you through the platform design, architecture and its use in several industry use cases as well as some of the underlying emotion detection technologies. A hackathon session at the end of the tutorial will enable you to work with the platform and try out some of the available components yourself.

Intended audience

Industry representatives and developers - providing insight into how emotion analysis can be used in business use cases and how the MixedEmotions platform for emotion analysis can be used in this