Trying to share some code ideas

logo

Month: February 2020

Lyrical Success

I’m not gonna write you a love songSara Bareilles

Post 1 of 4

A question I’ve always wondered is their a magic formula for creating a number one song. Many song writers are prolific however their song doesn’t necessarily have commercial success. It may be difficult to quantify how the musical composition relates to the song’s success. However I am going to make an attempt at evaluating the lyrics of songs to determining if we can accurately predict if a song will be commercially successful.

Lyrical Success – Getting the Data

To obtain the data I am going to use Beautiful Soup and a few other packages to scrap the content from the websites.

Part 2 of 4

Steps for Getting the Data

  1. Get the Songs Made by the Artists
  2. Extract the List of Songs by the Artist
  3. Scrap the Lyrics for Each Song by the Artist
  4. Extract the Lyrics from Each File
  5. Scrap Rankings by Artists
  6. Parse the Song Rankings Files

The first step is getting the list of the songs by the artist. I am using the website http://www.azlyrics.com to obtain the list of songs and the lyrics for the songs.

Lyrical Success – Preparing the Data

The data created in the previous step needs a little bit of cleaning up before we can get into the model building. It is a common step that needs to be undertaken to ensure that the data can be loaded into models without any issues.

Part 3 of 4

Steps for Preparing the Data

  1. Clean the Rankings
  2. Match the Song Ranks

Lyrical Success – Model Prediction

In this step, I will look at the data to see if we can do any feature engineering. And then I will edit the data for the model, train multiple models, evaluate the best model and then test the model. Let’s get started.

Part 4 of 4

Steps for Creating the Model

  1. Song Summary
  2. Visualizations
  3. Prepare the Lyrics for Analysis
  4. Model Building

Powered by WordPress & Theme by Anders Norén