Project Aim
This project seeks to prove that there are differences between the lyrics of different genres and we can use Data Science to show this.
Scraping Lyrics from Musixmatch.com
We will scrape lyrics from Musixmatch.com, a free website that hosts a tonne of accurate lyric content
Encoding lyrics using Google's Universal Sentence Encoder
We will represent lyrics using high dimensional vectors from Google’s Universal Sentence Encoder.
Reducing Dimensionality using PCA
We will apply Principal Component Analysis (PCA) to reduce these high dimensional representations
Visualising in Plotly
We will use Plotly and build a combination of simple plots and more sophisticated 2d and 3d scatter plots to see where there are similarities and differences in lyrics
