Abstract

Musical genres are constructed labels used to describe music that people tend to group together. As music becomes more accessible, consumers aim to expand their musical tastes more and more. Streaming services offer robust song recommendations, but fail to recommend genres. We attempt to implement various similarity and clustering methods to generate meaningful genre recommendations based off of a Spotify user's top tracks. Through our research, we explore methods of parameter optimization and novel preprocessing methods.

Advisor

Sommer, Nathan

Department

Computer Science; Mathematics

Disciplines

Databases and Information Systems | Data Science | Software Engineering

Keywords

Music, Genres, Recommendation System, Recommender System, Clustering, Database, Similarity, Spotify

Publication Date

2021

Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis

Download

Share

COinS
 

© Copyright 2021 Angelo Williams