Ece Yurdakul

Profile

Project title:

Emotion-based Personalised Music Recommendation    

Abstract:

The advent of music streaming services has brought about a fundamental shift in the way we consume music, making music recommendation a process supporting the listening experience. As the number of streaming options and the volume of available music continue to grow at an unprecedented rate, the need for sophisticated and personalized music recommendation systems that take into account individual preferences and listening habits has become increasingly important. While some research in music recommendation systems (MRSs) entirely ignores the impact of emotion, studies that do consider it generally rely on pre-tagged mood datasets. However, this approach assumes that all users have the same emotional response to a particular song, which is not the case . Emotional associations with music are listener-dependent and influenced by individual and hedonic factors such as culture, personality traits, and musical taste. 

The objective of this research project is to design, implement, and evaluate an emotion-based recommendation system that incorporates individual and hedonic factors in order to create more personalized song recommendations.

C4DM theme affiliation:

Music Informatics, Machine Listening

Research

Research Interests:

Recommendation, computational creativity, music generation