Google has quietly rolled out a new “Similar Items“ feature within Google Image Search on both the mobile web as well as Android‘s Google app, which uses machine learning to surface similar looking products matching what users are looking for. So far, this is limited to fashion and lifestyle products, such as sunglasses, handbags, or shoes, but Google says it will expand in the coming months to cover other apparel and home & garden categories. The feature will not only recognize what the objects in the image are, down to their brand and model but also find out about their price and availability. The feature also includes links to buy those similar-looking items on ecommerce sites.
What Brands Need To Do
This is an interesting play on Google’s part to build out its shopping-related features and explore how computer vision may help it breath new life into its search ads. This new feature revitalizes image search as a viable product discovery channel that brands will need to pay attention to. As users can also search by images, this can easily become a way for consumers to snap a picture of a physical product and quickly find similar items to buy online. While this feature currently runs on machine learning algorithms and is not monetized in any way, it is not hard to imagine how this could easily become a new ad product where online retailers selling the same items can bid to appear in the front end of the carousel.
Source: Marketing Land
Uber has acquired a New York-based startup Geometric Intelligence to launch its own in-house research team that focuses on artificial intelligence development, according to The New York Times. This move not only reaffirms Uber’s dedication to improve its algorithms for more efficient routing and ride-sharing, but also signals the company’s growing ambition in developing autonomous vehicles and driverless solutions, to which sophisticated AI and machine learning tools will be crucial.
Why Brands Should Care
As conversational interfaces and cloud-based solutions rise to prominence, AI and machine learning are quickly becoming a hot topic among the tech and ad industries, especially in regard to how they would transform the way we analyze data and extract insights. Already, we are seeing companies like IBM and Oracle integrating their respective machine learning-powered solutions into marketing products, promising profound impact on consumer expectations and brand-consumer interactions.
This is a topic we will be diving into in details in our upcoming Outlook 2017 report. Please check back in early January to read more of our take on this hot industry trend.
Source: The New York Times
Salesforce on Wednesday launched Commerce Cloud, a new marketing tool that integrates assets from Demandware that Salesforce acquired in June. Salesforce is integrating its CRM and other platforms with Commerce Cloud, which also comes with Apple Pay integration as well as predictive analytics and product recommendations powered by its “Einstein” artificial intelligence platform. At launch, the service is already powering 1,800 retail stores for brands such as Adidas, Lands’ End, and Pandora.
What Retailers Should Do
Retailers can use this new cloud-based marketing service to better serve and engage with shoppers. The Apple Pay integration enables retailers to offer one-touch checkout for both online and in-store purchases, creating a frictionless checkout experience while also closing the online-offline attribution loop. The integration with Einstein, on the other hand, adds some analytics and customization features into the mix to make the system more efficient and data-driven.
For more information on how retailers can better utilize customer data to connect with shoppers across various sales channel, check out the Boundless Retail section in our Outlook 2016.
Source: Marketing Land
Samba TV has acquired Filmaster, a startup based in Warsaw, Poland composed mostly of Data Scientists and developers or Artificial Intelligence software. Described as a service that offers content recommendations using “its own proprietary artificial intelligence and machine learning algorithms”, Filmaster will help boost Samba’s “capabilities in content recommendations and marketing automation”, per Samba TV’s press release.
As the world’s leading provider of Smart TV applications, Samba TV will soon begin incorporating Filmaster’s technology in its software so as to improve the relevance of cross-screen advertising, as Samba TV’s platform is monetized by brand sponsorships. Back in April, the Lab led our parent company Interpublic Group to invest in Samba TV, and we are excited to see what this acquisition may bring for them.
Header image courtesy of Samba TV
Read original story on: Wired
Last week, Amazon quietly launched a new service aimed at opening its own AI technology, which its mighty recommendation engine is built on, to all businesses to use. The new service, known as the Amazon Machine Learning Service, is designed to help developers easily integrate targeted recommendation engines based on data and machine learning into their own websites and apps. As part its ever-growing suite of AWS cloud computing services, Amazon continues to claim Internet infrastructures, one piece at a time.
A new system developed at the University of Rochester aims to give users genuine opinions about a restaurant before you commit your money. Called “nEmesis,” their software uses machine learning to listen to geotagged tweets that match a restaurant location. It follows that user’s tweets for 72 hours, and captures any information about them feeling ill. Though it’s not good at accounting for random bouts of the flu amidst genuine food poisonings, over a four month period it correctly identified 480 reports of food poisoning. So maybe before you risk that C-rated lunch spot in NYC, check out nEmesis to see what’s really going on.
All new features require an adoption period where users familiarize themselves with the tech. Graph Search is one of Facebook’s most ambitious initiatives that places Facebook at the center of online discovery, but also brings with it some serious user challenges. This article from Gigaom shows both the machine learning and human intervention required to streamline the product to gain mass adoption.
Google X creates 16,000-core ‘neural network’ for independent machine learning