Skip to content

cavitcakir/Predicting-Spotify-Top-List

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

Website: https://cs210groupproject.wordpress.com
If you are one of the people who is happy on sunny days and pessimistic and strangely calm on cloudy days, you are one of us. As like mood, we think music taste changes with weather, so does the most listened songs of a single day. In order to test this claim, we collect most listened 20 songs for a single day and a single location from Spotify. For example, we have the list the most listened 20 songs on Turkey on 01.01.2017. We enriched our data by collecting new attributes related to music such as its loudness, danceability, energy etc. Our motivation is to show human beings are more predictable than it thought, for example evena change weather conditions affects people’s music taste. Bear in mind, people are generallytoo arrogant when it comes to their music taste, showing how human beings (us) fragile to something as native as temperature is exciting. We are going to examine hypothesis like Does temperature affect loudness of a song, does temperature and energy of a together determine the rating of a song? Our final expectation is to predict a song’s likelihood of being in the hit list based on that day’s weather information with comparing previous relation between a song’s properties like loudness, danceability…