How Netflix Creates ‘Taste Communities’

Wired: “The Defenders provides Netflix with a unique case study. Instead of merely allowing it to find out if someone who likes, say, House of Cards also will like Daredevil (yes, BTW), it tells them which of the people who landed on Daredevil because of House of Cards will make the jump to The Defenders.”

“Wildly different programs lead people to The Defenders’ standalone shows. The top lead-in show for Luke Cage? Narcos. But for Iron Fist, it’s a Dave Chappelle special. Someone who watches Jones probably will watch Cage, but beyond that the groups of people—Netflix calls them ‘taste communities’—gravitating toward those shows enjoy very different programming.”

“Every Netflix user belongs to three or four taste communities. It’s easy to say that this influences what appears in your recommendations, but it’s not quite that simple. Membership in those communities does more than dictate the top 10 comedies appearing in a row of your queue, it determines whether comedies appear there at all … Each time you open Netflix it exposes you to 40 or 50 titles. Netflix considers it a win if you choose one of them.”


Disney Machine-Learns for Laughs

Quartz: Disney “is using machine learning to assess the audience’s reactions to films based on their facial expressions, it wrote in a new research paper. It uses something called factorized variational auto-encoders, or FVAEs, to predict how a viewer will react to the rest of a film after tracking their facial expressions for a few minutes.”

“The FVAEs learn a set of facial expressions, such as smiles and laughter, from the audience, and then make correlations between audience members to see if a movie is getting laughs or other reactions when it should be—a much more sophisticated version of how Amazon and Netflix make suggestions for new things to buy or watch based on your shopping or viewing history.”

“By placing four infrared cameras and infrared illuminators above a theater screen, the researchers were able to identify 16 million facial landmarks, or expressions, from more than 3,100 theatergoers during 150 screenings of nine Disney movies … the data was then analyzed with a computer. (Before this gets too creepy, Disney isn’t tracking your every move at your local theater. The experiment took place during screenings at one particular 400-seat theater. And audiences likely had to choose to participate.)”


Technology Cannot Hug a Customer

The New York Times: “Technology, some hotels are finding, has its limits. ‘Technology cannot hug a repeat guest,’ said George Aquino, the vice president and managing director of AHC+Hospitality … That is the reason his company, which manages several hotels, has been running a training program for some of its managers and other staff members to improve their hospitality skills, connect with local business leaders and learn more about local tourist offerings.”

“Similar programs are sprouting in other cities, involving not just hotels but also restaurants and even cities themselves, which see the personal touch as giving them a competitive edge. For business travelers, in particular, talking to someone knowledgeable about a city can lead to a good restaurant. And it can also help expand business leads.”

“A consulting program based in Tucson, Certified Tourism Ambassadors, trains hospitality workers. Mickey Schaefer, the chief executive and founder, said she had developed the idea in 2006 while working for the American Academy of Family Physicians to plan its conventions. Hospitality workers sometimes did not know their own cities, leading to bad experiences, she said … The program, she said, ‘is more than just helping the customer. It is helping them find the richness of whatever they are interested in.’ She added that the program also instills civic pride.”


Deep Thinking: ‘Artificial’ Trumps ‘Intelligence’

LARB: From a review of Deep Thinking, by Garry Kasparov, the chess champion defeated by Deep Blue, a machine, in 1997: “The history of computer chess is the history of artificial intelligence. After their disappointments in trying to reverse-engineer the brain, computer scientists narrowed their sights. Abandoning their pursuit of human-like intelligence, they began to concentrate on accomplishing sophisticated, but limited, analytical tasks by capitalizing on the inhuman speed of the modern computer’s calculations.”

“This less ambitious but more pragmatic approach has paid off in areas ranging from medical diagnosis to self-driving cars. Computers are replicating the results of human thought without replicating thought itself. If in the 1950s and 1960s the emphasis in the phrase ‘artificial intelligence’ fell heavily on the word ‘intelligence,’ today it falls with even greater weight on the word ‘artificial’ … If a machine can search billions of options in a matter of milliseconds, ranking each according to how well it fulfills some specified goal, then it can outperform experts in a lot of problem-solving tasks without having to match their experience or insight.”

Also: “A bit of all-too-human deviousness was also involved in Deep Blue’s win. IBM’s coders, it was later revealed, programmed the computer to display erratic behavior — delaying certain moves, for instance, and rushing others — in an attempt to unsettle Kasparov. Computers may be innocents, but that doesn’t mean their programmers are.”


Machine Platform Crowd: The Future Today

The Wall Street Journal: Machine Platform Crowd, a new book by Andrew McAfee and Erik Brynjolfsson, “is a book for managers whose companies sit well back from the edge and who would like a digestible introduction to technology trends that may not have reached their doorstep—yet … In the authors’ terminology, ‘Machine’ is shorthand for computers running software that, with new AI techniques called ‘deep learning,’ essentially teaches itself how to make judgments superior to those of humans. ‘Machine’ also encompasses the disappearance of employees in the services sector, leaving only the customer, robots and software—what the authors refer to as ‘virtualization.'”

“‘Platform’ refers to digital environments that bring economic actors together, exploiting free, or nearly free, online access, reproduction and distribution. Uber and Airbnb are examples of new platforms. ‘Crowd’ refers to information resources created by the uncredentialed, the nonexpert and, with rare exceptions, the unpaid. Wikipedia and the Linux operating system comprise the two most impressive achievements of the crowd.”

​”Messrs. McAfee and Brynjolfsson argue that, in the latest phase of the second machine age, incumbent businesses will be pushed aside if they fail to understand how new machines and software, platforms, and the crowd enlarge the scope of digital technologies—just as manufacturers that had appeared and thrived in the first phase of the first machine age were displaced when electricity supplanted steam power in the early 20th century.”


Algorithmania: Nutella Prints 7 Million Unique Labels

Hyperallergic: “Nutella’s manufacturer, Ferrero, recently partnered with advertising agency Ogilvy & Mather Italia to make Nutella even more endearing and its consumption more exciting by presenting Nutella Unica, an algorithm designed to create a series of unique labels for (almost) every Nutella jar in Italy. The algorithm pulls from a database of dozens of patterns and colors to create seven million different versions of the Nutella label — pink and green, striped and polka-dotted, Pop Art-inspired and minimal.”

“Advertising for Nutella Unica compares the individuality of each jar to the people of Italy themselves (there are about 60 million people in Italy, so about 11% of them can get a jar all their own — actually quite a feat). When these exceptionally delicious artworks hit shelves in Italy in February 2017, they sold out in barely a month. Can you imagine the shopping possibilities — buying multiples, or maybe trying to find an attractive pair?”


French Twist: How Yoplait Manufactures Authenticity

The New York Times: “Thick, sour Greek yogurts with names like Chobani, Fage and Oikos were surging in popularity. Sales of runny, sugary Yoplait were oozing off a cliff. So Yoplait executives ran to their test kitchens and developed a Greek yogurt of their own … They called it Yoplait Greek. It tanked almost immediately. And so has almost every other Greek yogurt product that Yoplait has put on shelves.”

“So now, Yoplait is opening a new front in the cultured-milk battles … They’re calling it Oui by Yoplait, in homage to the company’s French roots … if, as you are shopping, you happen to pick up a small glass pot of Oui and are momentarily transported to the French countryside, you’ll know that the company has finally figured out how to look beyond the data and embrace the narrative. Yoplait may have figured out how to fake authenticity as craftily as everyone else.”

“Yoplait began scouring its own history and ultimately found a tale that seemed to resonate: For centuries (or so the story goes), French farmers have made yogurt by putting milk, fruit and cultures into glass jars and then setting them aside. So Yoplait tweaked its recipe and began buying glass jars … It has a creamy texture and sweet flavor. And if this product is a success — if years from now someone tells the heartwarming story of how the Greek hordes were defeated by simple French pots — then we’ll know that Yoplait’s number crunchers finally figured out the formula for authenticity, and have reclaimed their crown.”


Quantumobiles: VW Catches D-Wave

The New York Times: “Efforts by Volkswagen, trying to remake itself as a technology leader as it recovers from an emissions scandal, show how far into exotic realms of technology carmakers are willing to go. Volkswagen, a German company, recently joined the handful of large corporations worldwide that are customers of D-Wave Systems, a Canadian maker of computers that apply the mind-bending principles of quantum physics.”

“While some experts question their usefulness, D-Wave computers — housed in tall, matte black cases that recall the obelisks in the science fiction classic 2001: A Space Odyssey — can in theory process massive amounts of information at unheard-of speeds … While classical computers are based on bits with a value of either 1 or 0, the qubits in a quantum computer can exist in multiple states at the same time. That allows them, in theory, to perform calculations that would be beyond the powers of a typical computer.”

“This year Volkswagen used a D-Wave computer to demonstrate how it could steer the movements of 10,000 taxis in Beijing at once, optimizing their routes and thereby reducing congestion … Such claims are met with skepticism by some experts, who say there is no convincing proof that D-Wave computers are faster than a well-programmed conventional supercomputer … Volkswagen executives say they will publish the results of their work with D-Wave computers, allowing outsiders to try to debunk them.”


‘Amazon Charts’ Re-Define ‘Best Seller’

The New York Times: Amazon now tracks “not only the top-selling digital and print books on Amazon, but the ones that customers spend the most time reading … With its lists, Amazon aims to redefine the notion of a best seller, expanding it to include books that are ‘borrowed’ from its e-book subscription service, and ones that are streamed on Audible. As a result, the lists give increased visibility to books that might not typically appear on other best-seller lists.”

“All of Amazon’s acquisitions and new features are having a cumulative effect, allowing the company to draw on its vast customer base and troves of data to discover what is popular, and return that information to customers, creating a lucrative feedback loop … Crowdsourcing and data mining are also driving the company’s approach to its bookstores, which act as showcases for books popular with customers on the site. While the stores have traditional categories, like fiction, nonfiction and travel, the most eye-catching shelves feature categories culled from Amazon’s customer data.”

“The first thing customers see when they walk into the store is a large display table, labeled Highly Rated, which includes books with an average rating of 4.8 stars or higher on a scale of 5 … Another display case, labeled Page-Turners, features books that people finish reading on their Kindle in fewer than three days … Another section features the most ‘wished for’ books from Amazon’s website … The books are all displayed face out. Under each book is a card with the average customer rating, the number of reviews and a featured review from an Amazon reader. Displaying the full cover of each book mimics the visual look of Amazon’s website, and might lure customers to unfamiliar titles.”


Google Eyes: Watch While You Shop

The Washington Post: “Google executives say they are using complex, patent-pending mathematical formulas to protect the privacy of consumers when they match a Google user with a shopper who makes a purchase in a brick-and-mortar store. The mathematical formulas convert people’s names and other personal information into anonymous strings of numbers.”

“The formulas make it impossible for Google to know the identity of the real-world shoppers, and for the retailers to know the identities of Google’s users, Google executives said. The companies know only that a match has been made. In addition, Google does not get a detailed description of the individual transactions, just the amount spent.”

“Google would not say how merchants had obtained consent from consumers to pass along their credit card information. In the past, both Google and Facebook have obtained purchase data for a more limited set of consumers who participate in loyalty programs. Consumers that participate in loyalty programs are more heavily tracked by retailers, and often give consent to share their data with third parties as a condition of signing up. (Not all consumers may realize they have given such consent, according to the digital privacy advocacy group Electronic Frontier Foundation).”