Understanding Music Consumption Patterns Through Lyrics

From Master Projects
Revision as of 12:46, 25 August 2018 by Sbl530 (talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

has title::Understanding Music Consumption Patterns Through Lyrics
status: ongoing
Master: project within::Cognitive Science
Student name: student name::Sven van den Beukel
Start start date:=01-05-2018
End end date:=30-09-2018
Supervisor: Albert Meroño Peñuela
Second supervisor: Victor de Boer
Company: has company::VU
Thesis: has thesis::Media:Thesis.pdf
Poster: has poster::Media:Posternaam.pdf

Signature supervisor



Abstract KIM 1

This study investigates whether, and to what extent, differences in the gender, nationality and age of users translate to differences in the contents of songs these users listen to. The focus lies primarily on extracting song topics and stylistic features (e.g. the number of unique - and function words used, or the fraction of lines that rhyme). Eventually, the proposed features are used to classify a user its age, gender and nationality respectively. Previously used features are added as well, so that the performance of the newly proposed features can be compared to previous research. Increasing the knowledge about the song contents that demographic user groups are more likely to be interested in, should help improve the accuracy of recommender systems on both serving new users and marketing new music pieces to their audience.