shutterstock_49233400

Love is All Around – Data Driven Digest

The joy of Valentine’s Day has put romance in the air. Even though love is notoriously hard to quantify and chart, we’ve seen some intriguing visualizations related to that mysterious feeling.  If you have a significant other, draw him or her near, put on some nice music and enjoy these links.

Got a Thing for You

We’ve talked before about the Internet of Things and the “quantified self” movement made possible by ever smaller, cheaper, and more reliable sensors.

One young engineer, Anthony O’Malley, took that a step further by tricking his girlfriend into wearing a heart rate monitor while he proposed to her. The story, as he told it on Reddit, is that he and his girlfriend were hiking in Brazil and he suggested that they should compare their heart rates on steep routes.

As shown in the graph he made later, a brisk hike on a warm, steamy day explains his girlfriend’s relatively high baseline pulse, around 100 beats per minute (bpm), while he sat her down and read her a poem about their romantic history.

What kicked it into overdrive was when he got down on one knee and proposed; her pulse spiked at about 145 bpm—then leveled off a little to the 125-135 bpm range, as they slow-danced by a waterfall.

Finally, once the music ended, the happy couple caught their breath and the heart rate of the now bride-to-be returned to normal.

A chart of the heart rate of programmer Anthony O'Malley's fiancée while he proposed. Image courtesy of Imgur.
A chart of the heart rate of programmer Anthony O’Malley’s fiancée while he proposed. Image courtesy of Imgur. View post on imgur.com.

 

What makes this chart great is the careful documentation. Pulse is displayed not just at 5-second intervals but as a moving average over 30 seconds (smoothing out some of the randomness), against the mean heart rate of 107 bpm.  O’Malley thoughtfully added explanatory labels for changes in the data, such as “She says SIM!” (yes in Portuguese) and “Song ends.”

Now we’re wondering whether this will inspire similar tracker-generated reports, such as giving all the ushers in a wedding party FitBits instead of boutonnieres, or using micro-expressions to check whether you actually liked those shower gifts.

Two Households, Both Alike in Dignity

One of the most famous love stories in literature, “Romeo and Juliet,” is at heart a story of teenage lovers thwarted by their families’ rivalry. Swiss scholar and designer Martin Grandjean illuminated this aspect of the play by drawing in a series of innovative network diagrams of all Shakespeare’s tragedies.

Network diagram of Shakespeare's "Romeo and Juliet," with kind permission of Martin Grandjean.
Network diagram of Shakespeare’s “Romeo and Juliet.” Published with kind permission of Martin Grandjean.

 

Each circle represents a character—the larger, the more important—while lines connect characters who are in the same scene together. The “network density” statistic indicates how widely distributed the interactions are; 100% means that each character shares the stage at least once with everybody else in the play.

The lowest network density (17%) belongs to Antony and Cleopatra, which features geographically far-flung groups of characters who mostly talk amongst themselves (Cleopatra’s courtiers, Antony’s friends, his ex-wives and competitors back in Rome). By contrast, Othello has the highest network density at 55%; its diagram shows a tight-knit group of colleagues, rivals, and would-be lovers on the Venetian military base at Cyprus trading gossip and threats at practically high-school levels.

The diagram of Romeo and Juliet distinctly shows the separate families, Montagues and Capulets. Grandjean’s method also reveals how groups shape the drama, as he writes:  “Trojans and Greeks in Troilus and Cressida, … the Volscians and the Romans in Coriolanus, or the conspirators in Julius Caesar.”

Alright, We’ve Got a Winner

Whether your Valentine’s Day turns out to be happy or disappointing, there’s surely a pop song to sum up your mood. The Grammy Awards are a showcase for the best — or at least the most popular — songs of the past year in the United States.

Musixmatch's machine learning model to predict Grammy-winning songs (Image by kind permission of Musixmatch)
Musixmatch’s machine learning model to predict Grammy-winning songs (Image by kind permission of Musixmatch).

 

The online lyrics library Musixmatch, based in Bologna, Italy, leveraged its terabytes of data and custom algorithms to make their prediction based on all 295 of the past Song of the Year nominees (going back to 1959).  As Musixmatch data scientist Varun Jewalikar and designer Federica Fragapane wrote, they built a predictive analytics model based on a random forest classifier, which ended up ranking all 5 of this year’s nominees from most to least likely to win.

Before announcing the predicted winner, Fragapane and Jewalikar made a few observations:

  • Song of the Year winners have been getting wordier, though not necessarily longer. (Most likely due to the increasing popularity of rap and hip-hop genres, where lyrics are more prominent.)
  • They’ve also been getting louder.
  • Lyrics are twice as important as audio.
Lyrics site Musixmatch charted the increasing volume of Grammy-nominated songs over the years. (Image by kind permission of Musixmatch.)
Lyrics site Musixmatch charted the increasing volume of Grammy-nominated songs over the years. (Image by kind permission of Musixmatch).

 

And they note that a sample set of fewer than 300 songs “is not enough data to build an accurate model and also there are many factors (social impact, popularity, etc.) which haven’t been modeled here. Thus, these predictions should be taken with a very big pinch of salt.”

With that said, their prediction… was a bit off but still a great example of visualized data.

Recent Data Driven Digests:

February 10: Visualizing Unstructured Content

January 29: Are You Ready for Some Football?

January 19: Crowd-Sourcing the Internet of Things

 

About Stannie Holt

Stannie Holt
Stannie Holt is a Marketing Content Writer at OpenText. She has over 20 years' experience as a journalist, market research analyst, and content marketing expert in the fields of enterprise business software, machine learning, e-discovery, and analytics.