Modelling music selection in everyday life with applications for psychology-informed music recommender systems
My research interests and current (PhD) project relate to the uses and functions of music in everyday life, Music Information Retrieval (MIR), and music recommender systems. I am specifically interested in why people select different music according to situational factors (e.g., activities, 2 locations) and as such am interested in understanding the functional and/or cognitive requirements of everyday situations that determine music selection. To this end, in my PhD, I have conducted psychometric research to identify and gauge functions of music listening in everyday listening situations and the ways in which these are influenced by situational factors, and by extension how these influence subsequent music selection.
Moreover, I am interested in how modern listening technologies (e.g., online streaming services) have changed the ways in which people access music, and how these technologies continue to curate listening experiences. This has led to an interest in music recommender systems in particular, and the methods applied in such systems to generate recommendations. I am therefore also interested in the uses of machine learning (particularly with behavioural data), but have also explored psychology-informed approaches to recommendations via explanatory modelling. Ultimately, I am interested in whether integrating knowledge about listening behaviours in predictive processes may mitigate the data dependency of existing systems on black-box machine learning models and generate more explainable recommendations to listeners/users in an age where trust and transparency in technology are crucial.