Visualizing the CORGIS Music Dataset

For this design sprint, we were tasked to create a website that uses visual methods to communicate data but for different end goals.

End Goals:

We decided to communicate data from the Music JSON Library from the CORGIS Dataset Project. This data set essentially contained information on “one million popular contemporary songs”. By “contemporary” the data set contains varying artists from the 60’s (like The Beatles) to today (like Jay-Z).

We started out on page 1, coming up with initial ideas of what we wanted to convey on our website. Some of our initial ideas revolved around visualizing song attributes based on an artist origin location, average tempo over the years, loudness vs tempo, and how much padding a song has from its fade in to fade out.

We then filtered our ideas to remove any that were too similar. After, we categorized them into what we liked and what we did not like. From this we came up with ideas for our three graphs for clear communication of the data.

On the original brainstorm design sheet, we had intended to compare location with artist, genre, and tempo/duration data. In the end however, we decided it would be more meaningful to just compare the number of songs and artists per origin state.

These will be represented on a color-coded map so they are easier to visualize.

For the persuasive communication we wanted to go with a visualization that resembled a sound visualizer, to create a memorable relative representation of the third map. The last part for this first page is to question what has been generated. We believe that this is a solution for the assignment because it communicates where in the United States certain aspects of music are popular.

The next step in the 5 Design sheet process is to come up with the initial designs.

We decided that our layout would be a webpage that had a map of the United States, a legend for the map, and buttons to switch the maps to the different data we want to show.

We wanted to show via color and shades how many songs, artists and average songs per artist are relative to each state. The user would operate the webpage by clicking on the button that they want to see the data for.

We found that the positives for this design is that having one map on screen at a time doesn’t overwhelm the user with information, and it is easier to see where certain location have better results. The negatives of this design is that we need to be careful with the colors we select for the map. If we select a poor choice such as red, those who are colorblind may have a hard time understanding the information.

The last step of the 5 design sheet process is to detail how we think the final build should look like. We found this to be very similar to our previous step’s page. The only differences is that we selected the color gradient to be green to black, green being the highest value and black being the lowest value. We also added information to the details box. The main ideas we listed are the database we are taking data from.

We were interested in the artists associated with these popular songs. For out first communication of data we displayed the American artists and where they are from on a map. The color varies state by state on a black to green scale, with black being the least number of artists from that state and green being the most.

We also measured where songs are being produced in the US. Similar to the artists per state, they are categorized from black to green with black being the least number of songs and green being the greatest number of songs.

Finally we also measured the average number of songs per artists per state. We did this by taking the number of songs released in that state and dividing it by the number of artists in that state. The scale for this is the same as the other maps with black being the lowest average and green being the highest average.

For the persuasion portion of the assignment we decided to display the average songs per artist in a more memorable way, to visualize in which states artists appear to produce more successful songs on average.

To do this, we had the specific states flash so that a higher bpm meant a more average songs per artist for that specific state.

Due to our time constraints, this one proved to be more of a proof of concept, and the representation of the data is a bit too abstract to create an understanding on the data on its own. Paired with the maps however, it caught viewers’ interest fairly effectively.

From the in class demos, The groups liked the map, but they did not like the colors we chose. They did not like the black to light green color scale because it was sometimes too hard to determine the difference between states with very similar values (although this was in part due to the low contrast on the monitor that we were demoing the website on).

We wanted to have a wider range of colors to be our scale but we were unable to get another color to work.

Also for the persuasion graph, some groups liked it and others did not. Some of the suggestions that we received that we would implement if we were to work on it further:

In addition, we would ideally give it a closer resemblance to an audio visualizer to make the visuals a bit cleaner and more organized.

Here you can use the buttons on top to select what data to see for the clear communication and to go to the beats per minute visualization for the persuasion data visualization.

Looking at our final prototype, its strengths is how the data is represented as a map, but its weakness is how the color scale makes it harder to see similar values. If we had more time, we would’ve liked to improve the color scale to make it more appealing and easy to read as well as make the website more appealing instead of having a white, blank background.

Related posts:

I have read a lot about how to deal with chronic pain and the takeaway message is to not allow pain to define us. I have tried this and not only does it not work for me, but it actually makes me feel…

I am not a religious person. The boundaries of religion feel too confining at times. This may seem surprising coming from someone who attended Catholic school from kindergarten through college, but…

Fecha en la que la boyband inglesa One Direction anunció su hiatus. Las fans siguen esperando una señal de su regreso, a pesar de que el tiempo del famoso “descanso” acabó. Se rumorea acerca de que…