In this article we focus on the personalization aspect of model building and explain the modeling principle as well as how to implement Photon-ML so that it can scale to hundreds of millions of users. Recommender systems are automated computer programs that match items to users in different contexts. Such systems are ubiquitous and have become an integral part of our daily lives. Examples include recommending products to users on a site like Amazon, recommending content to users visiting a website like Yahoo!, recommending movies to users on a site like Netflix, recommending jobs to users on LinkedIn, and so on. Given the significant heterogeneity in user preferences, providing personalized recommendations is key to the success of such systems.

Abode has revealed its Gen 2 gateway, and it's "100% committed" to bringing HomeKit support to the platform. More details will come "soon," the company said. However, this already makes it one of the more flexible hubs to date. It can already communicate with Amazon Alexa, Google Assistant and IFTTT, and it supports a raft of smart home devices using Z-Wave Plus, Zigbee and Abode's own abodeRF. Outside of HomeKit, Gen 2 is mainly notable for its inclusion of 4G cellular data backup on some models (if you subscribe to a plan) as well as the addition of Z-Wave Plus. It's clearly designed with home security in mind thanks a built-in battery for home security as well as a 93db siren.

After being roused from your sleep, you say "Good Morning" and the bathroom lights turn on immediately. As you get dressed for work in front of the mirror, it recognises who you are and starts displaying newsfeed customised for you. These scenarios were among the numerous artificial intelligence (AI) showcases at the latest CES in Las Vegas, the world's largest consumer electronics trade fair. From the pre-show presentation of tech sales and forecasts by CES 2019 organiser Consumer Technology Association (CTA) to the keynotes by different technology firms, AI was a topic constantly mentioned. Thomas Husson, vice-president and principal analyst at research firm Forrester, defines AI as "a system of capabilities for machines to interact, think or mimic human intelligence and engagement", which helps automate tasks and improves employee and customer experience.

We are tracking the most impactful AI use cases here. This is meant to be a list that grows over time so feel free to contribute with your comments, this list is definitely not comprehensive now. And share the knowledge with your twitter followers: @AndrewYNg claims that "AI is the new electricity". We compiled 100 applications runnning on this new electricity. Marketing can be summarized as reaching the customer with the right offer, the right message, at the right time, through the right channel while constantly learning. Optimizing product, pricing & placement allows marketers to create an attractive value proposition to customers. Gesture Control: Gesture control enables higher levels of activity and engagement by allowing users another mode of interaction with your digital products. Quantify the gesture levels and other engagements in order to provide meaningful insights. Pricing Optimization: Also called dynamic pricing or demand pricing, pricing optimization allows companies to optimize markdowns. Optimal markdowns minimize cannibalization while maximizing revenues. One of the easiest transformations a business can achieve, dynamic prices directly impact the bottom line and can be rolled out in a matter of days. Optimize markdowns to minimize cannibalization while maximizing revenues. Identify which products are of significant importance for customers.

Technology innovations have increasingly become boardroom discussions over the past decade. For the past couple of years, the conversation has revolved around artificial intelligence, and some companies have made heavy investments in technologies. While AI and machine learning remain hot topics, the big question now seems to be, "Where's the return on investment?" Your company can now analyze a lot of data and build compelling user profiles, but how does that manifest itself in achieving business goals? As the CEO of a customer experience solutions company, I advise companies on how to invest in chatbot technology as part of their digital transformation strategies.

Chatbots are finally having their day. This technology has been around for decades, but thanks to recent advances in artificial intelligence and natural language processing, these once annoying bots -- that could do little more than repeat back credit card numbers -- have transformed into incredibly useful and prolific communication tools. People from all 195 countries now use online chats to start conversations on business websites, and by 2020, more than 50% of medium to large enterprises will use chatbots to support products. Usage rates will continue to climb as chatbot technology steadily improves, allowing these digital assistants to answer more questions and to use more colloquial language to engage with humans. Despite these technological advances, companies are failing to leverage the full potential of these virtual digital assistants.

Collaborative filtering (CF) is the key technique for recommender systems (RSs). CF exploits user-item behavior interactions (e.g., clicks) only and hence suffers from the data sparsity issue. One research thread is to integrate auxiliary information such as product reviews and news titles, leading to hybrid filtering methods. Another thread is to transfer knowledge from other source domains such as improving the movie recommendation with the knowledge from the book domain, leading to transfer learning methods. In real-world life, no single service can satisfy a user's all information needs. Thus it motivates us to exploit both auxiliary and source information for RSs in this paper. We propose a novel neural model to smoothly enable Transfer Meeting Hybrid (TMH) methods for cross-domain recommendation with unstructured text in an end-to-end manner. TMH attentively extracts useful content from unstructured text via a memory module and selectively transfers knowledge from a source domain via a transfer network. On two real-world datasets, TMH shows better performance in terms of three ranking metrics by comparing with various baselines. We conduct thorough analyses to understand how the text content and transferred knowledge help the proposed model.

Conditional preference networks (CP-nets) are a graphical representation of a person's (conditional) preferences over a set of discrete features. In this paper, we introduce a novel method of quantifying preference for any given outcome based on a CP-net representation of a user's preferences. We demonstrate that these values are useful for reasoning about user preferences. In particular, they allow us to order (any subset of) the possible outcomes in accordance with the user's preferences. Further, these values can be used to improve the efficiency of outcome dominance testing. That is, given a pair of outcomes, we can determine which the user prefers more efficiently. Through experimental results, we show that this method is more effective than existing techniques for improving dominance testing efficiency. We show that the above results also hold for CP-nets that express indifference between variable values.

Amazon Alexa has been programmed to read the news headlines in the style of a newsreader. The popular voice assistant will now emphasise words, and mimic the intonation and pace of a TV anchor to present the news in a more natural way. Newsreader Alexa has been trained to read the daily bulletins when the user says'Alexa, what's the latest?' Amazon Alexa has been programmed to read the news headlines in the style of a newsreader. The virtual assistant already was able to read out the headlines but using the traditional robotic voice. Amazon conducted tests and found that people preferred hearing the news in this more realistic and listener friendly manner, compared to the robotic tone.

Artificial intelligence (AI) will have a major impact on the digital workplace in the very near future. IBM studies revealed that within the next three years about 120 million workers in the world's 10 largest economies may need to be retrained or reskilled because of AI and intelligent automation. IBM officials also found that 67 percent of CEOs believe AI will drive significant value in HR. But, how is AI impacting the digital workplace now? Gartner reported that AI is currently being applied to areas that include collaboration (fee required), content services, intranets, human capital management/recruiting, IoT in the digital workplace, help desk/IT service monitoring, knowledge management, meeting solutions, search and insight engines, and virtual employee assistants.