All activity on your social media accounts contributes to your “social graph,” which maps your interconnected online relationships, likes, preferred activities, and affinity for certain brands, among other things.
Now MIT spinout Infinite Analytics is leveraging these social graphs, and other sources of data, for very precise recommendation software that better predicts customers’ buying preferences. Consumers get a more personalized online-buying experience, while e-commerce businesses see more profit, the startup says.
The neat trick behind the software — packaged as a plug-in for websites — is breaking down various “data silos,” isolated data that cannot easily be integrated with other data. Basically, the software merges disparate social media, personal, and product information to rapidly build a user profile and match that user with the right product. The algorithm also follows users’ changing tastes.
Think of the software as a digital salesman, says Chief Technology Officer Purushotham Botla SM ’13, who co-founded Infinite Analytics and co-developed the software with Akash Bhatia MBA ’12. A real-world salesperson will ask consumers questions about their background, financial limits, and preferences to find an affordable and relevant product. “In the online world, we try to do that by looking at all these different data sources,” Botla says.
Launched in 2012, Infinite Analytics has now processed more than 100 million users for 15 clients, including Airbnb, Comcast, and eBay. According to the company, clients have seen around a 25 percent increase in user engagement.
Bhatia says the software also makes online-shopping searches incredibly specific. Users could, for instance, search for products based on color shade, textures, and popularity, among other details. “Someone could go [online] and search for ‘the most trending, 80 percent blue dress,’ and find that product,” Bhatia says.
Dismantling data silos
The two co-founders met and designed the software in course 6.932J (Linked Data Ventures), co-taught by Tim Berners-Lee, the 3Com Founders Professor of Engineering. Berners-Lee later joined Infinite Analytics as an advisor, along with Deb Roy, an associate professor of media arts and sciences, and Erik Brynjolfsson, the Schussel Family Professor of Management Science at the MIT Sloan School of Management.
As a class project, Bhatia and Botla, along with several classmates, designed software meant to dismantle data silos — a major theme in the class. “There’s so much data around us, but all the data is in silos, disconnected,” Botla says. “The goal was to take this data and make it more machine-readable and associate semantic meanings to it.”
But this first prototype wasn’t for finding products — it was for finding people. Looking at social media and other data, the software would determine the best way to reach out to a specific person, whether through mutual friends on LinkedIn or through other online channels. For instance, you could search for and find how to connect with any person who happens to, say, golf at a specific course.
With this idea, Bhatia and Botla launched Infinite Analytics and earned a semifinalist spot in the 2012 MIT $100K Entrepreneurship Competition. Later that year, they joined the inaugural class of MIT’s Founders’ Skills Accelerator (FSA) program, hosted in the Martin Trust Center for MIT Entrepreneurship — “which was pretty much a defining moment for us as a startup,” Bhatia says.
In the program, they realized that the code base could be best used to refine online shopping. “One thing led to another before we realized that we were sitting on a very powerful idea, which, if implemented right and executed well, could take us to being the next ‘unicorn,’” Bhatia says, referring to a startup with a valuation exceeding $1 billion.
By the end of the three-month FSA (now called Global Founders’ Skills Accelerator), Infinite Analytics had earned its first two e-commerce customers.
Profiling people and products
Today, the Infinite Analytics software operates in three steps that involve some complicated data crunching: creating a user profile, creating a product profile, and running both through a recommendation engine.
“So if you come to the website, we know who you are and, based on what product you’re looking at, we can show product most interesting to you with more precision than other recommendation software,” Botla says.
To create a user profile, the software captures all available user data — from the likes of Twitter, Facebook, LinkedIn, and Google Plus — which provides information on activities and interests, geographical location, employment history, media habits, and spending potential.
Using an analytics engine that includes natural language processing and other techniques, the software draws in-depth conclusions about those data. “That gives a very detailed view of the person,” Botla says.
At the same time, the software compiles a product profile, pulling information from product descriptions to detail the product’s function, and which activities it’s used for. Basically, this allows the software to accurately narrow down products that a user will buy.
If a user, for example, had recently gone from being a student spending $100 online to a business professional or CEO of a well-funded startup, the software will factor in that the user’s spending potential has increased and recommend a more expensive item. If that user happened to prefer baseball to soccer, the software would then offer all baseball-related products, in that price range, while dismissing any that were related to soccer.
Another innovation is the software’s use of image processing to break down the image of a product into quantifiable components of color, shape, patterns, and texture — “making the product information embedded in images machine readable to be consumed by algorithms,” Botla says.
In February, the company tested this image-processing proficiency by analyzing the viral phenomenon that had social media users hotly debating whether a dress was white and gold or blue and black. Using a low-resolution image of the dress, the Infinite Analytics software quantified the exact colors: about 48 percent of two shades of blue, 39 percent of three shades of black, and 3 percent gold. (As it turns out, the dress was indeed blue and black.)
All this means that if a user was, for instance, looking at a shirt with dark blue and orange stripes in a certain pattern, the Infinite Analytics software could, for example, match a consumer with a different piece of clothing with that exact pattern. “That’s the power of the platform we’ve built,” Bhatia says.
