Keynotes

Abstract: Musicologists, neuroscientists, philosophers, psychologists have long puzzled over the nature of music-evoked emotion and have addressed, among other things, the question of whether or how music can actually trigger emotional states. In this address, I will give an overview of our research on the characterization and classification of music-induced emotions, highlighting some of the tools that were developed in this process, notably the Geneva Emotion Music Scale (GEMS) and the Emotion-to-Music-Mapping- Atlas (EMMA). I will then turn to what we have learned about factors that demonstrably promote or restrict the experience of music-evoked emotions, including dispositional and situational ones. Finally, I will briefly outline some of the possibilities and challenges in applying this work to AI-assisted analysis of music-evoked emotions.

Bio: Marcel Zentner studied Psychology, Psychopathology and Philosophy at the University of Zurich. After receiving his PhD, he held professional positions at Harvard University, UC-Berkeley, the University of Geneva, and the University of York (GB). In 2013 he was appointed professor of personality psychology and assessment at University of Innsbruck. Zentner’s research interests are in the areas of personality development, psychological assessment, the psychology of mating, emotion and music. His research has been published in the premier natural science journals Nature and the PNAS as well as in the leading psychology journals such as the Journal of Personality and Social Psychology and Psychological Science. He is also the author of the Handbook of Temperament (Guilford Press), coedited with R. Shiner. Zentner’s research has been featured by BBC ONE, National Public Radio, The Wall Street Journal, among others, and has received funding though the ESRC, the Swiss National Science Foundation, the Jacobs Foundation, and Google. 

Abstract: One of the descriptions of recommender systems is that they are information retrieval systems where the query is the user. However, unlike stable, objective queries the user is a complex subject, whose traits, attitudes, feelings, and behaviour are interwoven and hard to model. In this talk I will provide a framework for inferring psychology-informed characteristics and using them in recommender systems.

Bio: Marko Tkalcic is associate professor at the Faculty of Mathematics, Natural Sciences and Information Technologies (FAMNIT) at the University of Primorska in Koper, Slovenia. He aims at improving personalized services (e.g. recommender systems) through the usage of psychological models in personalization algorithms. To achieve this, he uses diverse research methodologies, including data mining, machine learning, and user studies.