Determinantal point processes (DPPs) are an elegant model for encoding probabilities over subsets, such as shopping baskets, of a ground set, such as an item catalog. They are useful for a number of machine learning tasks, including product recommendation. DPPs are parametrized by a positive semi-definite kernel matrix. Recent work has shown that using a low-rank factorization of this kernel provides remarkable scalability improvements that open the door to training on large-scale datasets and computing online recommendations, both of which are infeasible with standard DPP models that use a full-rank kernel. In this paper we present a low-rank DPP mixture model that allows us to represent the latent structure present in observed subsets as a mixture of a number of component low-rank DPPs, where each component DPP is responsible for representing a portion of the observed data. The mixture model allows us to effectively address the capacity constraints of the low-rank DPP model. We present an efficient and scalable Markov Chain Monte Carlo (MCMC) learning algorithm for our model that uses Gibbs sampling and stochastic gradient Hamiltonian Monte Carlo (SGHMC). Using an evaluation on several real-world product recommendation datasets, we show that our low-rank DPP mixture model provides substantially better predictive performance than is possible with a single low-rank or full-rank DPP, and significantly better performance than several other competing recommendation methods in many cases.

This book is ideal for anyone wishing to apply graph algorithms to Big Data, esp. It targets experienced software developers, but reviews the basics of Spark and the Scala language as well as introducing GraphX. The pages are packed with information presented concisely and the book moves right along. The book is very readable and personable (I think the style rather enjoyable to read), but I found it helpful to look up certain terms or acronyms online for more information on unfamiliar topics that the authors sometimes mention in passing. The many charts and graphs were well done and particularly helpful in making the material understandable.

BENGALURU: Customer engagement plays an important part in traditional retail outlets which seems to be missing among the plethora of e-commerce portals. To resolve this, a bunch of young entrepreneurs have come up with an Artificial Intelligence-driven sales chatbot which engages customers with a human touch. And the good news is that it's live and will provide round-the-clock hands-on solutions and technical support. Navneet Gupta, Chief Technical Officer of Racetrack.ai We understand that customer engagement is of utmost priority for any business and we have designed it to not only intelligently engage with the customer but also make the conversations fruitful with every chat." He points out that the AI-driven platform can not only convert a simple query to a customer but can also attain a loyal customer for life by a unique customer engagement."

Apple Inc. acquired artificial intelligence startup Turi Inc. for about 200 million, according to people familiar with the situation, in the latest deal by the iPhone maker to accumulate advanced computing capabilities for its products and services. Turi helps developers create and manage software and services that use a form of AI called machine learning. It also has systems that let companies to build recommendation engines, detect fraud, analyze customer usage patterns and better target potential users, according to the Seattle-based startup's website. Apple could use this to more rapidly integrate the technology with future products. Apple's move Friday is part of a broader battle among Google, Facebook Inc. and Amazon.com

Artificial intelligence enables Staples to automate ordering and customer service through its Easy Button. Staples Inc. is testing technology enabling business customers to order products by voice via its Easy Button. The office supplies retailer is applying machine learning technology to the button, allowing customers to press it to order or reorder a product by voice, or to ask common order-related questions, such as when an order will be delivered or the status of a return. The move is part of a big push by Staples Inc., No. 21 in the Internet Retailer 2016 B2B E-Commerce 300, to use machine learning to automate ordering and customer service, says Ryan Bartley, director of mobile for Staples. Machine learning refers to computer programs that teach themselves to grow and change when exposed to new data, without being programmed by an individual.

One-to-one customer engagement enabled by machine learning yields a 13 percent reduction in bounce rate and a 33 percent average order value improvement at Marmot. A recent study from Gallup shed light on just how valuable it is to invest in customer engagement. Fully engaged customers, defined as those who have had a measurable reaction, connection, or experience with your brand, represent a 23 percent share of wallet, profitability, revenue, and relationship growth premium compared to average shoppers. By contrast, the study found that disengaged customers -- those who have no emotional connection to your brand -- represent a 13 percent discount in those same measures. It's no surprise that customer engagement and loyalty applications have been hot in retail, particularly e-commerce, where customer data is easier to gather and easier to analyze than in any other channel.

Online giant Amazon said it profit in the second quarter surged ninefold to 857 million, lifted by cloud services as the tech giant expanded its offerings. Revenues jumped 31 percent to 30.4 billion, Amazon said in stronger-than-expected results. While Amazon developed a reputation for delivering little or no profit in its retail operations, its earnings have been growing over the past year as it expands into video and new delivery services and boosts its cloud computing unit, Amazon Web Services (AWS). Revenues jumped 31 percent to 30.4 billion, Amazon said in stronger-than-expected results The second annual Prime Day was the biggest day ever for Amazon, and was also a record day for Amazon devices globally. Compared to Prime Day 2015, worldwide orders grew by more than 60%, orders from third-party sellers with Prime Day deals nearly tripled, and Prime members saved over twice as much on deals.

Apple Inc., AAPL 0.23 % stepping up its involvement in one of Silicon Valley's hottest arenas, is acquiring machine-learning specialist Turi Inc. Financial terms couldn't immediately be determined. GeekWire, which reported the acquisition Friday, said Apple is paying around 200 million. An Apple spokesman issued the company's standard statement in connection with such transactions: "Apple buys smaller technology companies from time to time, and we generally do not discuss our purpose or plans." Turi's chief executive is Carlos Guestrin, who holds the title of Amazon professor of machine learning at the University of Washington--a position endowed by Amazon.com He is also an associate professor of computer science and engineering.

Machine-learning is all the rage in fraud detection, with industry analysts, academics, businesses and technology media examining the advantages of algorithms and big data in the fight against e-commerce fraud. Especially for fraud analysts working in companies with small budgets, machine-learning tools are seen as a cost-effective way to tighten fraud controls while maintaining fast decision times, as Forrester noted in its 2015 cross-channel fraud report. There's no question that machine-learning tools can be an effective component of fraud reduction program, but relying on them to save staffing costs may not be cost-effective in the long run. That's because while machine learning is an invaluable tool in the fight against fraud, it relies on human input and insight to create a comprehensive solution that yields the best results. Algorithms are useful for identifying potential fraud quickly, but due to variability in consumer behavior – such as making online purchases while traveling abroad -- some transactions will be falsely flagged for decline.

Macy's is testing a mobile tool using artificial intelligence that lets shoppers get answers customized to the store they're in -- such as where a particular brand is located or what's in stock -- that they would normally ask a sales associate face-to-face. The tool, which the nation's largest department store chain calls a "mobile companion," can be accessed for now through a browser and will accept questions in 10 U.S. locations about products, services and facilities. It uses natural language and offers feedback in seconds. It's developed by IBM Watson, the Jeopardy-winning "cognitive computing" service and is designed to keep learning more about the store's customers. That's a key element as Macy's seeks to spur sluggish sales, make being at the store more enjoyable and distinguish itself from online portals and specialty retailers.