Emotion-Based Music Recommendation from Quality Annotations and Large-Scale User-Generated Tags

Marta Moscati, Hannah Strauß, Peer-Ole Jacobsen, Andreas Peintner, Eva Zangerle, Marcel Zentner, Markus Schedl

Abstract

Emotions constitute an important aspect when listening to music. While manual annotations from user studies grounded in psychological research on music and emotions provide a well-defined and fine-grained description of the emotions evoked when listening to a music track, user-generated tags provide an alternative view stemming from large-scale data. In this work, we examine the relationship between these two emotional characterizations of music and analyze their impact on the performance of emotion-based music recommender systems individually and jointly. Our analysis shows that (i) the agreement between the two characterizations, as measured with Cohen’s κ coefficient and Kendall rank correlation, is often low, (ii) Leveraging the emotion profile based on the intensity of evoked emotions from high-quality annotations leads to performances that are stable across different recommendation algorithms; (iii) Simultaneously leveraging the emotion profiles based on high-quality and large-scale annotations allows to provide recommendations that are less exposed to the low accuracy that algorithms might reach when leveraging one type of data, only.
BibTeX
@inproceedings{moscati2024emotion,
  author = {Marta Moscati and Hannah Strauß and Peer-Ole Jacobsen and Andreas Peintner and Eva Zangerle and Marcel Zentner and Markus Schedl},
  title = {{Emotion-Based Music Recommendation from Quality Annotations and Large-Scale User-Generated Tags}},
  booktitle = {UMAP},
  year = {2024},
  doi = {10.1145/3627043.3659540},
}

Loading related content...

Figure