Spotify High-Popularity Tracks - Exploratory Data Analysis

An in-depth exploratory data analysis of Spotify's high-popularity tracks, uncovering patterns in audio features, genre distributions, and temporal trends that define commercially successful music.

---

Key Findings

Metric	Value
Tracks Analyzed	1,686
Unique Artists	1,437
Genres Covered	6
Time Span	1957 – 2024

• Popular tracks average 0.67 energy, 0.65 danceability, and 0.53 valence — a balanced mix of intensity and accessibility
• Strongest correlation: Energy and Loudness (r = 0.68); strongest negative: Energy and Acousticness (r = -0.58)
• Major keys dominate at 58%, with C#/Db as the most common key
• Pop, rock, and hip-hop account for nearly half of all high-popularity tracks
• Modern releases (2020s) overwhelmingly dominate popularity charts

---

Visualizations

The notebook generates 20 publication-quality visualizations, including:

#	Visualization	Description
01	Audio Feature Distributions	Histograms with KDE for all 9 audio features
02	Feature Box Plots	Comparative box plots on a normalized 0–1 scale
04	Correlation Heatmap	Pearson correlation matrix of audio features
07	Genre Radar Chart	Normalized audio profiles for top 6 genres
08	Genre Violin Plots	Distribution shape comparisons across genres
11	Feature Evolution	Audio feature trends across decades
17	Pair Plot	Pairwise relationships colored by genre
18	2D Density Plot	Energy vs. Danceability density distribution
19	Cluster Map	Hierarchical clustering of audio features
20	Summary Dashboard	Combined overview of all key findings

---

Dataset

Two CSV files sourced from Spotify's audio features API:

• high_popularity_spotify_data.csv — 1,686 tracks (popularity score >= 68)
• low_popularity_spotify_data.csv — available for comparative analysis

Audio Features

Feature	Range	Description
energy	0.0 – 1.0	Intensity and activity measure
danceability	0.0 – 1.0	Suitability for dancing
valence	0.0 – 1.0	Musical positiveness / happiness
acousticness	0.0 – 1.0	Acoustic vs. electronic
speechiness	0.0 – 1.0	Presence of spoken words
instrumentalness	0.0 – 1.0	Likelihood of no vocals
liveness	0.0 – 1.0	Presence of live audience
loudness	-60 – 0 dB	Overall loudness in decibels
tempo	~50 – 210	Beats per minute (BPM)
track_popularity	0 – 100	Spotify popularity score

---

Getting Started

Prerequisites

• Python 3.8+
• pip

Installation

# Clone the repository
git clone https://github.com/your-username/spotify-eda.git
cd spotify-eda

# Create and activate a virtual environment
python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Requirements

numpy>=1.21.0
pandas>=1.3.0
matplotlib>=3.4.0
seaborn>=0.11.0
scipy>=1.7.0
jupyter>=1.0.0

Running the Notebook

jupyter notebook archive/spotify_eda_analysis.ipynb

Or open directly in VS Code with the Jupyter extension.

Output

After running all cells, the notebook produces:

• 20 PNG visualizations in the archive/ directory
• processed_spotify_data.csv — cleaned and feature-engineered dataset

---

Project Structure

spotify-eda/
├── README.md
├── requirements.txt
├── archive/
│   ├── spotify_eda_analysis.ipynb      # Main analysis notebook
│   ├── high_popularity_spotify_data.csv # Primary dataset
│   ├── low_popularity_spotify_data.csv  # Comparison dataset
│   ├── PROJECT_GUIDE.md                # Detailed code walkthrough
│   ├── processed_spotify_data.csv      # Generated after running
│   └── *.png                           # Generated visualizations

---

Analysis Sections

1. Data Loading & Exploration — Shape, types, missing values, duplicates
2. Data Cleaning & Feature Engineering — Datetime extraction, key mapping, duration conversion
3. Descriptive Statistics — Central tendency, spread, skewness, kurtosis
4. Distribution Analysis — Histograms, KDE, box plots, popularity breakdown
5. Correlation Analysis — Heatmap, scatter plots, top correlated pairs
6. Genre Analysis — Distribution, radar charts, violin plots, popularity comparison
7. Temporal Analysis — Release trends, decade evolution, monthly patterns
8. Musical Key & Mode Analysis — Key distribution, major/minor split
9. Artist Analysis — Top artists, audio feature profiles
10. Advanced Visualizations — Pair plots, 2D density, hierarchical clustering

---

Tech Stack

Library	Purpose
NumPy	Numerical operations and array computations
Pandas	Data manipulation, cleaning, and aggregation
Matplotlib	Static visualizations and custom plot layouts
Seaborn	Statistical visualizations and enhanced aesthetics
SciPy	Skewness, kurtosis, and statistical tests
Jupyter	Interactive notebook environment

---

Future Work

• Comparative analysis with low-popularity tracks
• Predictive model to forecast track popularity
• K-means clustering to identify track archetypes
• Time-series forecasting of audio feature trends
• Interactive dashboard with Plotly or Streamlit

---

License

This project is licensed under the MIT License. See LICENSE for details.

---

Acknowledgments

• Dataset sourced from the Spotify Web API
• Built with Python's data science ecosystem