- Your title can change over time.
Details for Milestone are available on Canvas (left sidebar, Course Project).
Our group is interested in the underlying patterns that contribute to the popularity of music. Despite the wide variety of tastes in society, there are likely trends that explain what factors permeate the most listened to and celebrated songs. Our goal then, is to determine what some of these elements could be and their relative influence. Some of the broader questions that could inspire more specific research include: 1. Which distinct elements are most valuable in popular music? 2. Is there a combination of elements that is present in the most popular music? 3. Are these same factors missing in the least popular music? This last question particularly aims to tackle the possible assumption in the previous two, that only musical elements are relevant to the success of music. If for example, we discover that the least popular music has similar characteristics to the most popular music, then perhaps we could suggest that more social factors or even the identity of artists are (to a degree) more important than the features of the music itself. I could imagine an interesting spotify music themed dashboard, that uses this dataset to reveal to the public what features of music they most value.
The dataset used in our project is sourced from Kaggle and observes several features on 13,024 unique songs. The purpose of the data collection was to build regression models that can be used to predict the song popularity. The dataset assesses different music based on song name, popularity, duration, acoustics, danceability, energy, instrumentals, key, liveness, and loudness. The sample songs used are part of a large range of music across different genres and historical time frames which will allow us to understand the popularity of music in different contexts. While the date of collection is not listed explicitly, the data was last updated about a year ago. The dataset can also be narrowed down by looking at the youngest song in the catalogue during our exploratory data analysis. While we speculate that this dataset will help us further our understanding of what makes music popular, there are several limiting factors that are important to acknowledge. Firstly, the dataset is relatively large which might make it harder to conduct our analysis. Additionally, the factors being observed about the songs have shown strong multicollinearity, which increases the complexity of the dataset.
The dataset used in our project is sourced from Kaggle and observes several features on 13,024 unique songs. The purpose of the data collection was to build regression models that can be used to predict the song popularity. The dataset assesses different music based on song name, popularity, duration, acoustics, danceability, energy, instrumentals, key, liveness, and loudness. The sample songs used are part of a large range of music across different genres and historical time frames which will allow us to understand the popularity of music in different contexts. While the date of collection is not listed explicitly, the data was last updated about a year ago. The dataset can also be narrowed down by looking at the youngest song in the catalogue during our exploratory data analysis. While we speculate that this dataset will help us further our understanding of what makes music popular, there are several limiting factors that are important to acknowledge. Firstly, the dataset is relatively large which might make it harder to conduct our analysis. Additionally, the factors being observed about the songs have shown strong multicollinearity, which increases the complexity of the dataset.
- Alvin Nganga: I am a bodacious bro who loves music and efficient teamwork!
- Tobi Ogunbote: Hi! I love to cook and work with amazing people.
- Annabelle Ngarambe: I am interested in popular culture, and I am excited to create cool visualizations.
{You should use this area to add a screenshot of an interesting plot, or of your dashboard}
{Add your stuff here}