This proposal states the intended work to be performed by spo-R-ify Group (member names noted above) for the final project of Data Analytics Essential course.
Our data obtained directly from Spotify Web API. For API connection, we created “Client ID” and “Client Secret” from Spotify for Developers Website. Thus, we can access instantly audios, artists, users and user playlists data. For further analysis, “spotifyr” package will be used.
Each song is separated within genres, and it’s easy to feel it if a song clearly reflects its emotion. So is it possible to describe the emotion of each song using its features such as energy and valence?
How do two different playlists differ between emotions?
Curious about all this, we shaped our project. We decided to create emotion maps of songs and music playlists in line with the features of the songs in the data tables that we will obtain using the Spotify WEB API.
We will present the mood analysis of certain songs, artists and albums by creating plots and graphs using Shiny.
Do countries differ from each other according to the music they listen to? How?
We will make inferences and comparisons on certain countries’ musical preferences, visualizing their mood attributes, by using Shiny R on local musical charts.
We will build a Shiny app, where users can enter their (or their loved ones’) playlists on Spotify, and see their “musical mood/personality” map. There will be twelwe pre-defined musical horoscope signs (made-up according to the mood attributes of the playlists), and their definitions to be returned to the user as the output of the app along with their mood analysis plots.
Spotifyr package, returns us variables such as, tempo, duration, key of the songs and Spotify’s audio analysis features including danceability, acousticness,energy, insturmentalness, liveness. We are planning to create new concepts like ‘Musical Horoscope’ with the help of the Spotify’s audio analysis variables and feature ggplot graphs that visualise the analysis data. Briefly,with the help of Spotifyr and Shiny packages, we intend to write a Shiny app, where user enters his/her playlist id, and as an output, the Shiny app presents musical anaylsis of the content, including concepts like ‘Musical Horoscope’, ‘Emotional Attributes of the playlists’.