Visualizing Ludwig van Beethoven’s Musical Journey

Ludwig van Beethoven was a German composer, who is one of the most popular composers in classical music history. The main inspiration behind this project was in honor of Beethoven’s 250th birth anniversary celebrated in 2020. Beethoven’s compositions are mainstays of Western classical music and are performed regularly by symphony orchestras around the world.

• Data Acquisition (Spotify API) : The dataset was accessed using the Spotify API using the spotipy package. The Spotify API returns an artist’s metadata in JSON format and this has been transformed into a Pandas dataframe for easier data analysis.

• Data Storytelling : Visual analysis has been performed using Matplotlib and Seaborn packages.

• Dataset Information : This dataset has all the compositions from Beethoven’s repertoire. The observations describe several audio features of each track including valence, acousticness, loudness,etc. Detailed information about the descriptive features can be found here Audio features of tracks.

Initial Overview

This dataset primarily consists of quantitative values that describe the musical qualities of each composition. Let’s begin with a basic exploration of some of the key features.

• Distribution of three key features in the dataset.

Valence is the positiveness of a track, tempo is the speed of the track and key measures the key the track is in, where the pitch is mapped to integers.

Valence	Tempo	Key

• Distribution of features over the duration of the tracks.

Most of the compositions are relatively short, as evident from the graphs below.

Loudness	Valence

• Pairwise scatterplots and jointplots.

Tempo & Loudness	Valence & Tempo	Valence & Danceability

• Features Heatmap : This heatmap displays the correlation between features. As evident from the image, most of the features are positively correlated with one another, albeit with a low correlation number.

• 3-D Scatterplots : Visualizing the interaction between three features in the same graph.