Song Genre Classification Based On Lyrics

CHOONG, ZE BIN (2020) Song Genre Classification Based On Lyrics. [Final Year Project] (Submitted)

[thumbnail of 23717_Choong Ze Bin.pdf] PDF
23717_Choong Ze Bin.pdf
Restricted to Registered users only

Download (3MB)


Text mining is often associated with the process of the English language to derive meaningful information or discover the underlying trend about a group of text content. There are many techniques that can be used to perform text mining and categorization is one of the technique. In this paper, a studyxon thexsong genre classificationxbased on the lyrics will be conducted. Songs can be categorized in many ways but generally those parameters are usually related to the musical aspect of the song. Extracting information from the song lyrics to classify it makes the most sense as the lyrics are what truly expressing the context of the song. The project’s main objective is to come up and verify the classification models that can be used to classify song by genre based on the lyrics. The models must posses the ability to analyze the song lyrics and classify it into one of the five selected popular genres: country, pop, rock, hip hop and rhythm and blues (R&B). The project will focus on several classification models such as Bernoulli Naive Bayes, Multinomial Naive Bayes, Random Forest, Decision Trees and K-nearest-neighbor to classify each and every song in the data set. The song dataxwill be extractedxusingvweb scrappingxtool and the lyrics will bexobtained through anxAPI invPython. After the pre processing and training phase, the accuracy of each classification model will bevanalyzed and compared tovdeterminevthe best model to be used to classify songs into genres based on the lyrics.

Item Type: Final Year Project
Subjects: Q Science > Q Science (General)
Departments / MOR / COE: Sciences and Information Technology > Computer and Information Sciences
Depositing User: Mr Ahmad Suhairi Mohamed Lazim
Date Deposited: 23 Sep 2021 23:45
Last Modified: 23 Sep 2021 23:45

Actions (login required)

View Item
View Item