/Spotify-Songs-Analysis

Analysis and Hypothesis Testing of Spotify Songs Dataset

Primary LanguageJupyter Notebook

Top 10000 Songs on Spotify Dataset

Contents

Data Cleaning

  • Null values handling
  • Duplicates removal

Data Understanding and EDA

  • Histogram
  • Boxplot
  • Bar chart
  • Time analysis
  • Correlation analysis

Hypothesis Tests

  1. What is the distribution of Danceability feature?

  2. Can we say that Time signature feature has a binomial distribution?

  3. How do Energy and Danceability relate to each other?

  4. How do Energy and Loudness relate to each other?

  5. What is the most effective feature on Popularity?

  6. We had seen before in the table that 'Energy' and 'Danceability' have nearly equal means. Are they really equal or is it just by chance?

  7. Can we say the distribution of 'Energy' and 'Danceability' is the same?