In this project, our main objective is to conduct an exploratory data analysis (EDA) on a dataset of the 100 most streamed songs on Spotify of all time. We will leverage the power of several popular Python libraries, including Numpy, Pandas, Matplotlib, Seaborn, and Beautifulsoup, to help us analyze and visualize the data effectively. Our goal is to identify any patterns or correlations that may exist within the data by creating and examining different graphs and charts. By performing this EDA, we aim to gain a better understanding of the musical features and trends that have made these songs so popular among Spotify users. In this report, we will present our findings and provide a detailed explanation of the tools and techniques we used throughout the analysis.
Based on the insights gathered, it is suggested that an ideal popular song should possess several characteristics:
- It should have moderate danceability, meaning that it should be easy to dance to, but not too fast-paced.
- Its tempo should fall between 90-100bpm.
- It is also important for the song to convey positive emotions. This can be achieved by using uplifting lyrics and a cheerful melody.
- It is recommended that the song should have less acoustic elements. This could involve using more electronic or synthesized sounds.
- Should have a length between 2-4 minutes.
- It should contain vocals and lyrics, which are important in creating a connection between the artist and the listener.
- Additionally, the song should be recorded in a major key, as this is often associated with positive emotions.
- It should have high energy, which can be achieved through the use of fast-paced rhythms, energetic vocals, and lively instrumentation.
These findings can be useful for music producers and artists in creating more commercially successful songs.