Spotify EDA: Data-Driven Music Discovery

---

Spotify Dataset EDA

This project performs an Exploratory Data Analysis (EDA) on a Spotify dataset to uncover insights about streaming patterns, artist performance, and audio feature correlations.

---

Project Structure

• spotify_dataset.csv – Raw dataset used for analysis.
• spotify_eda.ipynb – Jupyter Notebook with step-by-step EDA and visualizations.
• spotify_eda.pdf / spotify_eda.html – Exported report of the EDA with insights and plots.

---

Setup & Requirements

To run this project, install the following Python libraries:

pip install pandas numpy matplotlib seaborn

---

Analysis Steps

1. Data Loading & Cleaning

   • Removed unnecessary ID and index columns.
   • Converted numeric-like columns (e.g., Streams, Artist Followers) into numeric types.
   • Handled missing values using mean, mode, and median imputation.
   • Engineered features such as Release Year, Total Chart Presence, and Duration (min).

2. Feature Engineering

   • Created a Mood Score using valence and acousticness.
   • Adjusted zero values where invalid by replacing them with medians.

3. Exploratory Visualizations

   • Distribution of Streams (log-scaled).
   • Popularity vs. Streams scatterplot.
   • Correlation Heatmap of audio features.
   • Top 10 Artists by Total Streams bar chart.
   • Song Duration Distribution with mean highlighted.
   • Top Genres by Average Streams.
   • Yearly Trends in Average Streams.
   • Outlier detection (Streams & Popularity).

---

Key Insights

• Song Duration: Most tracks are ~3–4 minutes, aligning with commercial standards.
• Top Artists: Taylor Swift, BTS, and Justin Bieber dominate total streams.
• Distribution: Streaming counts follow a long-tail effect – few mega-hits drive most of the volume.
• Popularity: Most songs fall in the 60–90 popularity range, but niche or new tracks can still chart.
• Audio Features: High-energy, danceable songs are more likely to achieve streaming success.

---