Project2

Goal

Assess music genre trends over the past century and predict which genre of music will be the most listened to in the upcoming year.

Software and Platform

Types of Software Used:

• Programming Language: Python
• Data Processing & Analysis:
  • Pandas - Reading, cleaning, filtering, and merging the dataset
  • Facebook Prophet - Time series forecasting of genre popularity
• Data Visualization: Matplotlib & Seaborn Creating bar charts, line graphs, and other custom visualization aesthetics
• Data Storage & Export: CSV Files
• Cloud Computing: Google Colab

Packages Installed:

• pandas
• numpy
• matplotlib
• seaborn
• prophet

Platforms Used:

Windows & Mac

Map of Documentation

graph TB
  A[Project2]
  A1[README.md]
  A2[LICENSE.md]
  B[SCRIPTS]
  B1[EDA.ipynb]
  B2[cleaning merged data.ipynb]
  B3[genre predictive analysis.ipynb]
  B4[genre trend analysis.ipynb]
  B5[merging tracks and albums.ipynb]
  C[DATA]
  C1[cleaned_merged_tracks.csv]
  C2[raw_albums.csv]
  C4[obtaining tracks data.md]
  C3[Data Appendix]
  D[OUTPUT]
  D1[Analysis Output.pdf]
  D2[EDA.pdf]

  A --> A1
  A --> A2
  A --> B
  B --> B1
  B --> B2
  B --> B3
  B --> B4
  B --> B5
  A --> C
  C --> C1
  C --> C2
  C --> C3
  C --> C4
  A --> D
  D --> D1
  D --> D2

  %% Styling for main project folder (Dark Blue)
  style A fill:#003366,stroke:#001f3f,stroke-width:2px,color:white;

  %% Styling for main categories (Light Blue)
  style A1 fill:#4a90e2,stroke:#003d5b,stroke-width:2px,color:white;
  style A2 fill:#4a90e2,stroke:#003d5b,stroke-width:2px,color:white;
  style B fill:#4a90e2,stroke:#003d5b,stroke-width:2px,color:white;
  style C fill:#4a90e2,stroke:#003d5b,stroke-width:2px,color:white;
  style D fill:#4a90e2,stroke:#003d5b,stroke-width:2px,color:white;

  %% Styling for subsets (Light Red)
  style B1 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style B2 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style B3 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style B4 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style B5 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style C1 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style C2 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style C3 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style C4 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style D1 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;
  style D2 fill:#ff9999,stroke:#8b0000,stroke-width:1px,color:black;

Loading

Reproduction Instructions

1. Set up Environment

   • install python
   • install required libraries (listed above)

2. Download Dataset

   • navigate to DATA folder
   • download raw_albums.csv
   • follow instructions in obtaining tracks data.md to download tracks.csv

3. Clean Data and EDA

   • navigate to SCRIPTS folder
   • open merging tracks and albums.ipynb
   • run each cell to:
     • load in raw_albums.csv and tracks.csv
     • clean dataset
     • merge and download the datasets into one called merged_tracks_and_albums.csv
   • navigate to SCRIPTS
   • open cleaning merged data.ipynb
   • run each cell to:
     • load in merged_tracks_and_albums.csv
     • clean dataset
     • download cleaned version- cleaned_merged_tracks.csv
   • naviagate to SCRIPTS
   • open EDA.ipynb
   • run each cell to:
     • load in cleaned_merged_tracks.csv
     • perform EDA
     • visualize genre trends

4. Analyze Genre Trends

   • navigate to SCRIPTS
   • open genre trend analysis.ipynb
   • run each cell to:
     • load in cleaned_merged_tracks.csv
     • perform time series decomposition
     • generate visualizations of genre trends

5. Genre Predictions

   • navigate to SCRIPTS
   • open genre predictive analysis.ipynb
   • run each cell to:
     • load in cleaned_merged_tracks.csv
     • use Prophet model to predict future genre listens
     • visualize predictions