Print Email Facebook Twitter Understanding the relationship between user emotion and latent musical features Title Understanding the relationship between user emotion and latent musical features Author Shastry, Aishwarya (TU Delft Electrical Engineering, Mathematics and Computer Science) Contributor Tintarev, Nava (mentor) Houben, Geert-Jan (graduation committee) Liem, Cynthia (graduation committee) Degree granting institution Delft University of Technology Programme Computer Science | Data Science and Technology Date 2019-08-23 Abstract With the advent of Internet and resulting data boom, Recommender Systems have come to rescue by filtering the information available on the internet by providing us with relevant information. These systems come handy when one wants to listen to songs, watch movies or even buy products on the Internet. Primarily, these recommender systems used content based or collaborative filtering techniques to recommend items. More recent research has studied the importance of contextual features in recommender systems. Music preference has always been associated with the contextual feature emotion. However, few studies study the mood congruence effect in the domain of music recommender systems. The field of music emotion recognition also remains unexplored with recommendations being made with limited features. \\ This master thesis analyses the relationship between few latent musical features and user emotion through our interface MooDify. It is a music recommendation system that incorporates emotion in a user using emotion induction techniques and investigate the effect of their emotional state on satisfaction and unexpectedness when presented with songs curated to specific musical features. To achieve this, we analysed the enjoyment and unexpectedness ratings for recommendations specific to latent musical features for a given emotional state. We have been able to achieve some interesting results through this study which has been discussed later in this work. Subject Music recommenderEmotionMusical features To reference this document use: http://resolver.tudelft.nl/uuid:06248c28-894a-4c69-9978-d2982a35fc9a Part of collection Student theses Document type master thesis Rights © 2019 Aishwarya Shastry Files PDF Master_Thesis_Aishwarya_Shastry.pdf 2.18 MB Close viewer /islandora/object/uuid:06248c28-894a-4c69-9978-d2982a35fc9a/datastream/OBJ/view