Scalable real-time assortment optimization has become essential in e-commerce operations due to the need for personalization and the availability of a large variety of items. While this can be done when there are simplistic assortment choices to be made, imposing constraints on the collection of feasible assortments gives more flexibility to incorporate insights of store-managers and historically well-performing assortments. We design fast and flexible algorithms based on variations of binary search that find the revenue of the (approximately) optimal assortment. In particular, we revisit the problem of large-scale assortment optimization under the multinomial logit choice model without any assumptions on the structure of the feasible assortments. We speed up the comparisons steps using novel vector space embeddings, based on advances in the fields of information retrieval and machine learning. For an arbitrary collection of assortments, our algorithms can find a solution in time that is sub-linear in the number of assortments and for the simpler case of cardinality constraints - linear in the number of items (existing methods are quadratic or worse). Empirical validations using the Billion Prices dataset and several retail transaction datasets show that our algorithms are competitive even when the number of items is $\sim 10^5$ ($100$x larger instances than previously studied).

Our favorite affordable smart doorbell is on sale right now, and it's perfect for anyone looking for more peace of mind. Purchases you make through our links may earn us a commission. Wish there was an easier way to see who's at your door without having to go to all the trouble of getting up and checking it out for yourself? Smart doorbells are the way to solve that problem, as they deliver handy alerts, video feed, and two-way audio in a way designed to really compliment your already tricked out smart home. Smart doorbells tend to be rather pricey (see our top pick, the Google Nest Hello Smart WiFi Video Doorbell, which retails for $229 at Walmart).

Artificial intelligence (AI) is "completely living up to its expected hype and hysteria," with 70% of daily digital interactions being AI-based, and businesses generating multimillion-dollar revenue streams from it. This was the word from Mike Bugembe, founder of UK-based AI consultancy, Lens.ai, delivering a keynote at the ITWeb Business Intelligence Summit 2020, in Johannesburg, today. Bugembe is a bestselling author, international speaker and executive advisor, helping organisations use data and AI to transform their businesses and grow. Discussing the importance of an AI strategy to gain business value, Bugembe noted that companies across the globe are ramping up investments in AI-related technologies and gaining multimillion-dollar-revenue streams, cutting costs, managing risk, improving operations, and finding innovative ways to develop products and strengthen customer intimacy. However, he warned that without an intelligent roadmap, companies risk focusing on the wrong opportunities, resulting in failure to tap into the true promise of AI. "Business and technology experts believe AI will be the most significant technological revolution that businesses have ever experienced," said Bugembe.

Have you ever wished that you had someone who would guide you through your shopping list, or give you an informed suggestion on what to buy for any particular purpose? In today's day and age given we have so many options and advertisements cluttering our minds with unnecessary trivia about various products which we may or may not need, it is important to think beyond this. Product plug-ins are becoming increasingly common for various platforms like YouTube and Instagram to name a few. While they are able to target the audience much better than the more conventional modes of advertisement like television or newspapers, they still lack that personal touch. Now imagine an AI-based system that will be able to address this problem by establishing a personalized database.

The way Americans shop has changed drastically over the past 10 years and traditional brick and mortar shops have been hit hard by this. In 2019 alone it has been expected that more than 8,600 stores would close continuing a trend that coined the term retail apocalypse. But all is not lost, Artificial Intelligence is making a big impact on these businesses that could save the retail industry from dying. In just two years retail use of AI grew by 600%. Fifteen percent of companies reported spearheading the investment and adoption of AI systems and 1 in 4 large retailers are using their IT budget to invest up to 10% on AI systems.

What was the last thing you bought online? Why didn't you go to a store to buy it? Was it for the vast number of items you could scroll through before making a choice or the recommendations that the website or app pulled up for you, remembering your choices and interests from a previous visit, or the ease with which you paid for it with a card whose details were already stored with the website? The online shopping experience is leaps and bounds ahead of the traditional experience in terms of using data and technology to provide unique and personalized customer experiences. While brick and mortar stores also have their own upsides, the move towards omni-channel retailing today is key.

The pursuit of data-driven decision-making can make business leaders starry-eyed about data science, believing that artificial intelligence in particular can instantly transform their business. What's needed is a healthy tension between data scientists and business leaders around what's possible and workable for using data to drive key decisions. The ideal scenario is all parties in complete alignment. This can be envisioned as a perfect rectangle, with business leaders' expectations at the top, fully supported by a foundation of data science capabilities -- for example, when data science and AI can achieve management's goal of reducing customer retention costs by automating identification and outreach to at-risk customers. Consider Target, which in the mid-2010s had flat in-store sales and a growing digital presence.

Retail today looks completely different than it did five years ago, and five years from now it will look completely different than it does today. Technology is advancing at a pace that requires retailers to not just keep pace with these changes, but stay ahead of the adoption curve in order to remain competitive and top of mind for consumers. The past five years have focused largely on the sophistication of omnichannel retail. This essentially entailed putting nice wrappers around a number of backend technologies to present a frictionless user experience to the customer. However, the next five years will be defined by unified commerce or bringing all disjointed systems together into one system of record that provides cohesiveness and visibility across systems.

Amazon CodeGuru (Preview) is a service launched at AWS re:Invent 2019 that analyzes the performance characteristics of your application and provides automatic recommendations on ways to improve. It does this by profiling your application's runtime (with CodeGuru Profiler) and by automatically reviewing source code changes (with CodeGuru Reviewer). For more information, see What Is Amazon CodeGuru Profiler? This post gives a high-level overview of how CodeGuru Profiler works, common ways to use it, and how to improve your understanding of your application's performance in production. It assumes a basic knowledge of the JVM (Java Virtual Machine) and related concepts such as threads and call stacks. CodeGuru Profiler provides insights into your application's runtime performance with a continuous, always-running production profiler.

Supermarkets need to ensure clean and safe environments for both shoppers and employees. Slips, trips, and falls can result in injuries that have a physical as well as financial cost. Timely detection of hazardous conditions such as spilled liquids or fallen items on supermarket floors can reduce the chances of serious injuries. This paper presents EdgeLite, a novel, lightweight deep learning model for easy deployment and inference on resource-constrained devices. We describe the use of EdgeLite on two edge devices for detecting supermarket floor hazards. On a hazard detection dataset that we developed, EdgeLite, when deployed on edge devices, outperformed six state-of-the-art object detection models in terms of accuracy while having comparable memory usage and inference time.