Spotify Data App Overview

The Spotify Data App provides an interactive platform where users can explore musical data, including song sentiments, musical attributes like energy and danceability, and popularity metrics. It allows users to filter tracks based on genres and visualize various insights through interactive plots.

In the data preprocessing step, the dataset undergoes several stages, including handling missing values, data reduction, and converting columns into appropriate formats for analysis. The app performs sentiment analysis using VADER on track names and visualizes the relationships between musical attributes and sentiments.

Data Preprocessing and Reduction

Before diving into the sentiment analysis and visualization, the dataset was carefully preprocessed:

1. Duplicates Removed: Duplicate rows were eliminated to ensure the dataset was clean.

2. Handling Missing Data: Missing values in important columns like user_id and product_id were dropped, while missing values in rating_count were filled with 0.

3. Data Sampling: The dataset was reduced to 50% of its original size to make it more manageable for analysis.

4. Data Type Conversions: Columns representing prices (both actual and discounted) were cleaned and converted to numeric types for analysis.

5. String to Float Conversion: The rating column, which was initially in string format, was converted into numeric values for better analysis.

6. Sentiment Analysis: Sentiments (positive, neutral, and negative) were generated based on track names using the VADER sentiment analyzer. This provides insight into the emotional tone conveyed by the song titles.

By the end of this process, a clean and reduced dataset, ready for visualization and analysis, was created and saved as reduced_dataset.csv.

Dashboard and Interactive Visualizations

Sentiment Distribution by Genre

An interactive bar chart shows how sentiment varies across genres. Users can explore which genres contain more positive, neutral, or negative tracks, offering insights into the tone associated with different genres.

Sentiment vs. Popularity

This visualization uses a box plot to show how the sentiment of a track affects its popularity. Users can see if positive, neutral, or negative sentiments tend to have higher popularity scores.

Sentiment vs. Danceability & Energy

These visualizations explore the relationship between song sentiment and musical attributes such as danceability and energy. For example, are upbeat and danceable songs more likely to have positive sentiments?

Mood Navigator

A scatter plot allows users to find songs based on their preferred mood metric, such as energy or danceability, and compare it to the song’s sentiment and popularity. This helps users discover songs that fit their current mood or energy level.

Data Visualization Techniques

1. Vibrant Color Palette:

   • Positive Sentiment: #00CC96 (green)

   • Neutral Sentiment: #636EFA (blue)

   • Negative Sentiment: #EF553B (red)

   The use of contrasting colors makes it easier for users to differentiate between sentiments.

2. Clean and Intuitive Layout:

   • Each plot is designed with a clean white background, and the color scheme is consistent throughout the dashboard to provide a seamless experience. Titles and axes are clearly labeled for better comprehension.

3. Interactive Tabs:

   • The dashboard is divided into tabs (e.g., “Mood Navigator,” “Sentiment vs. Popularity”), allowing users to explore different insights in an organized manner.

Results and Insights

By visualizing the Spotify dataset, the following insights can be observed:

• Sentiment Distribution: The majority of tracks have neutral sentiments, with fewer tracks classified as positive or negative.

• Impact of Sentiment on Popularity: While there are neutral tracks with high popularity, positive sentiment may also correlate with higher popularity in some genres.

• Exploring Moods: The Mood Navigator provides a useful way to find songs that match a user's mood based on attributes like energy and danceability, which can be highly personal preferences.

This app combines robust data preprocessing, sentiment analysis, and interactive visualizations, providing a comprehensive way for users to explore their music data in a meaningful and engaging way.