If you\u2019ve shopped in the online world, you\u2019ve encountered recommendation engines. These artificial intelligence (AI) systems, also known as recommendation systems or recommender systems, leverage algorithms that help users find products and services based on their past buying behaviors, known preferences and more. Through their ability to predict interests and desires at a personalized level, recommendation engines help content and product providers drive people to music, video, books, clothes and just about any other product or service they might be interested in.\nServices like Amazon, Netflix, Spotify and YouTube make heavy use of recommendation engines in an effort to increase sales and improve customer satisfaction. But that\u2019s just the start of the use cases for these AI-driven systems. Shopping for a new audio system? Best Buy has some recommendations tailored to your tastes. Need to recruit a new member for your team? LinkedIn has a talent search and recommendation system for that. Looking for a new friend or partner? There are a lot of recommendation engines for that pursuit \u2014 namely, sites like Match.com and eHarmony.\n\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 ____________________________________________________________________\nAI subsets \n\nMachine learning is a subset of AI where algorithms that take data, learn from the data and then use that information to make informed decisions when presented with a new set of data. The software is always learning, taking in new information and thus, making better decisions over time. \nDeep learning is a subset of machine learning where additional information may be used to make decisions. Where a machine learning system is trained for a particular task and may require engineer guidance over time, deep learning systems are able to associate various data with the decision that must be made. _______________________________________________________________________\n\nUse cases\nIn the business world, the use cases for recommendation engines are both broad and deep. Let\u2019s look at a few examples.\nRetail\nGoogle, one of the pioneers in the field, offers a new service called Recommendations AI that makes it easy for retailers to adopt their own recommendation engines. This service draws on Google\u2019s years of experience in delivering recommended content across the company\u2019s flagship properties, including Google Ads, Google Search and YouTube.\nRecommendations AI enables retailers to build personalized recommendation systems based on\u00a0 leading-edge deep learning models \u2014 and, importantly, without the need for expertise in machine learning or recommendation systems. Recommendations AI dynamically adapts to customer behavior in real time, changing variables like assortment, pricing and special offers to drive improvements in customer engagement, conversion rate and basket size.1\u00a0\nThe importance of systems like these in the retail world is underscored in a new report from IDC. It notes that the worldwide retail industry will invest more than $5 billion in AI systems in 2019, and nearly half of that amount will go toward automated customer service agents, expert shopping advisors and product recommendation systems.2 Clearly, when it comes to these AI systems, retailers are seizing the day.\nMedia\nIntel worked with Cloudera to design a content-recommendation engine for a Chinese language media company. This system automatically creates a personalized news portal for each known reader and then recommends personally relevant articles. This personalization encourages visitors to read more articles and spend more time on the company\u2019s website.\nOver time, the company can further enhance the user experience by providing more responsive content and innovation products and services that evolve along with readers\u2019 changing behaviors and preferences. As an Intel case study notes, this AI-driven engine is helping the company grow its readership by providing fresh content \u2014 including news, videos and interactive games \u2014 based on each reader\u2019s personal interests.3\u00a0\nMarketing\nEpsilon is the digital powerhouse behind the marketing and loyalty programs of many Fortune 500 companies, including American Express, FedEx and Walgreens. As a Dell EMC case study notes, to help its clients build closer customer relationships, Epsilon uses AI and machine learning to analyze customer activity and deliver personalized content, email and recommendations based on individual preferences.\nWhen you consider the scale at which Epsilon operates, the company\u2019s work becomes even more amazing. Epsilon processes billions of emails per week with AI-driven systems that combine analytical expertise and predictive models to evaluate responses and create personalized messages in real time. As the company\u2019s CIO explains, \u201cEverything we do is centered around data and our ability to get the right message to the right person at the right time.\u201d4\u00a0\u00a0\nThe IT fuel for the recommendation engine\nTo make recommendation engines work in real time and at scale, companies need high performance computing clusters, complete with high-speed storage systems, that can process massive amounts of data in real time. And there\u2019s good news on this front, because the HPC infrastructure needed to run these AI-driven applications is getting much easier to deploy.\nHere\u2019s one example: The new Dell EMC Ready Solution for AI \u2013 Deep Learning with Intel delivers a ready-to-go solution for the development of AI-driven applications, including recommendation engines. It provides an optimized solution stack that simplifies the entire workflow, including all the hardware, software and services needed to get an AI solution up and running quickly.\nTo help with the demands of data-driven AI applications, Deep Learning with Intel provides documented and supported integration of Dell EMC Isilon storage to accelerate the movement of data. Isilon All-Flash scale-out NAS is designed to deliver the analytics performance and extreme concurrency at scale to consistently feed the most data-hungry analytic algorithms and eliminate I\/O bottlenecks.\nKey takeaways\nThe bottom line here should be pretty clear. Recommendation engines are powerful tools for building closer customer relationships and driving higher sales volumes across a wide range of use cases and industries. And today we have the IT fuel we need to power these engines \u2014 in the form of ready-to-deploy deep learning solutions.\nNow, we just need to fire up these solutions and hit the AI accelerator.\nTo learn more\n\nFor a closer look at Epsilon\u2019s successes with a AI-driven recommendation engine, read the Dell EMC case study \u201cGlobal Marketing Company Fuels Growth with AI.\u201d\nTo explore leading-edge HPC solutions for powering AI-driven applications, visit Dell EMC Ready Solutions for AI.\n\nAdvancing the frontiers of AI\nDramatic advances in data analytics and high performance computing capabilities have created a foundation for the adoption of AI-driven applications in the enterprise. However, these enabling technologies are only part of the AI story. The other part is the rise of smarter algorithms that can glean insights from massive amounts of data. In this series of posts, we explore these building blocks for AI solutions in enterprise environments.\n\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad\u00ad______________________\n\nGoogle, \u201cRecommendations AI (Beta),\u201d accessed September 9, 2019.\nIDC, \u201cWorldwide Spending on Artificial Intelligence Systems Will Be Nearly $98 Billion in 2023, According to New IDC Spending Guide,\u201d September 4, 2019.\nIntel, \u201cIntel and Cloudera Help Design a Content Recommendation Engine for Chinese Company,\u201d April 2016.\nDell EMC, \u201cGlobal Marketing Company Fuels Growth with AI,\u201d August 2018.