Honors Theses

Date of Award

Spring 5-1-2021

Document Type

Undergraduate Thesis

Department

Mathematics

First Advisor

John Latartara

Second Advisor

Gerard Buskes

Third Advisor

Michael Worthy

Relational Format

Dissertation/Thesis

Abstract

The study of music recommender algorithms is a relatively new area of study. Although these algorithms serve a variety of functions, they primarily help advertise and suggest music to users on music streaming services. This thesis explores the use of linear discriminant analysis in music categorization for the purpose of serving as a cheaper and simpler content-based recommender algorithm. The use of linear discriminant analysis was tested by creating lineardiscriminant functions that classify Wilco’s songs into their respective albums, specifically A.M., Yankee Hotel Foxtrot, and Sky Blue Sky. 4 sample songs were chosen from each album, and song data was collected from these samples to create the model. These models were tested for accuracy by testing the other, non-sample, songs from the albums. After testing these models, all proved to have an accuracy rate of over 80%. Not being able to write computer code for this algorithm was a limiting factor for testing applicability on a larger-scale, but the small-scale model proves to classify accurately. I predict this accuracy to hold on a larger-scale because it was tested on very similar music when in reality, these models work to classify a diverse range of music.

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.