Data mining has changed the way we think about information. Machine-learning algorithms now routinely chomp their way through data sets of Twitter conversations, travel patterns, phone calls, and health records, to name just a few. And the insights this brings is dramatically improving our understanding of communication, travel, health, and so on.
But there is another historical data set that has been largely ignored by the data-mining community—photographs. This presents a more complex challenge.
For a start, the data set is vast, spanning 150 years since the dawn of photography. What’s more, the information it contains can be hard to distill, often because it is too complex or too mundane to describe in words.
Today, that changes thanks to the work of Shiry Ginosar at the University of California, Berkeley, and a few pals, who have pioneered a machine-vision approach to mining the data in ordinary photographs.
These guys start with a relatively simple database—American high-school yearbook photographs dating back to 1905. These yearbook photos have been digitized on large scale by local libraries all over the U.S. and show full frontal photos of individuals in a standard pose.
Ginosar and co downloaded over 150,000 of these portraits. After removing those that were not full frontal portraits, they were left with some 37,000 images from more than 800 yearbooks from 26 U.S. states.
They then grouped the portraits by decade and superimposed the images to produce an “average” face for each period. This process revealed other “average” features for each period such as hairstyle, clothing, style of glasses, and even average facial expressions. The image above shows these averages for each decade for men and women.
The results make for interesting reading. A particularly striking feature is the evolution of smiling in yearbook photographs. Ginosar and co say that in the years after the invention of photography, most people adopted the same pose they would have used for a painted portrait—a neutral expression that would be easy to hold for a long period.
“Etiquette and beauty standards dictated that the mouth be kept small—resulting in an instruction to “say prunes” (rather than cheese) when a photograph was being taken,” say Ginosar and co.
But that changed during the 20th century, when photography became more popular. In particular, the photography company Kodak used advertising to popularize the idea of smiling in photos so that the images recorded happy memories.
Whatever the reason, smiling has become much more prominent. “These days we take for granted that we should smile when our picture is being taken,” say Ginosar and pals.
And the data backs that up. The team developed an algorithm for determining the degree of lip curvature in the photographs and this showed a clear trend in increasing smile intensity over time.
The data also reveals another trend. “Women significantly and consistently smile more than men,” they say. This is not a new discovery—indeed it has been discussed for decades.
to read the full article.