Sparsity of the User-POI matrix is a well established problem for next POI recommendation, which hinders effective learning of user preferences. Focusing on a more granular extension of the problem, we propose a Joint Triplet Loss Learning (JTLL) module for the Next New ($N^2$) POI recommendation task, which is more challenging. Our JTLL module first computes additional training samples from the users' historical POI visit sequence, then, a designed triplet loss function is proposed to decrease and increase distances of POI and user embeddings based on their respective relations. Next, the JTLL module is jointly trained with recent approaches to additionally learn unvisited relations for the recommendation task. Experiments conducted on two known real-world LBSN datasets show that our joint training module was able to improve the performances of recent existing works.

Cold start is an essential and persistent problem in recommender systems. State-of-the-art solutions rely on training hybrid models for both cold-start and existing users/items, based on the auxiliary information. Such a hybrid model would compromise the performance of existing users/items, which might make these solutions not applicable in real-worlds recommender systems where the experience of existing users/items must be guaranteed. Meanwhile, graph neural networks (GNNs) have been demonstrated to perform effectively warm (non-cold-start) recommendations. However, they have never been applied to handle the cold-start problem in a user-item bipartite graph. This is a challenging but rewarding task since cold-start users/items do not have links. Besides, it is nontrivial to design an appropriate GNN to conduct cold-start recommendations while maintaining the performance for existing users/items. To bridge the gap, we propose a tailored GNN-based framework (GPatch) that contains two separate but correlated components. First, an efficient GNN architecture -- GWarmer, is designed to model the warm users/items. Second, we construct correlated Patching Networks to simulate and patch GWarmer by conducting cold-start recommendations. Experiments on benchmark and large-scale commercial datasets demonstrate that GPatch is significantly superior in providing recommendations for both existing and cold-start users/items.

Were you unable to attend Transform 2022? Check out all of the summit sessions in our on-demand library now! Marketers know customers don't just crave personalized content -- they expect it in the business-to-consumer (B2C) and business-to-business (B2B) brands with which they interact. When trying to execute highly personalized email campaigns, however, marketers encounter many barriers. More than 40% of marketers cite a lack of resources -- time, people, and money -- as a significant challenge.

In the very near future, Amazon's famed voice assistant, Alexa, may sound quite different from the dutiful (and impersonal) voice you've grown accustomed to since it rolled out in 2014. At least, that's what Rohit Prasad, Amazon's senior vice president and head scientist for Alexa, announced at Amazon's re:MARS conference, a global artificial intelligence (AI) event that Amazon founder and executive chair Jeff Bezos hosted over the summer. With just a one-minute audio sample, the technology could bring a loved one's voice bounding through an Echo device's speakers. Prasad used a short presentation to show the audience how the new speech-synthesizer technology could help us forge lasting memories of our deceased relatives. "Alexa, can grandma finish reading me The Wizard of Oz?" A young boy asked a cute Echo speaker with big Panda eyes.

Google Home no longer needs to lean solely on smart home devices like thermostats to know whether or not you're around. Home's optional presence sensing feature can now use interactions with Nest speakers and smart displays to help detect activity in your abode, letting it perform automated actions. If you talk to your Nest Audio or tap your Nest Hub, for instance, Google may know to turn the lights on. Second-gen Nest Hubs can also use their Soli radar sensor to tell when you're close. You can enable presence sensing in the Google Home app for Android and iOS by visiting the Features section in the settings.

In the past, brick-and-mortar retailers leveraged native marketing and advertisement channels to engage with consumers. They have promoted their products and services through TV commercials, and magazine and newspaper ads. Many of them have started using social media and digital advertisements. Although marketing approaches are beginning to modernize and expand to digital channels, businesses still depend on expensive marketing agencies and inefficient manual processes to measure campaign effectiveness and understand buyer behavior. The recent pandemic has forced many retailers to take their businesses online.

Amazon's Echo Show smart displays with Alexa voice control are already a good value next to the competition, but a bunch of deals are making them even cheaper. The Show 5 is the best one, on sale right now for just $40, or 53 percent off the regular price. And if you want a larger screen, you can find great deals on the Show 8 and Show 15 -- including a free Show 5 with the latter. The Echo Show 5 scored a very solid 85 score in our Engadget review, as it's small size is ideal if don't have a ton of space on your desk, nightstand or countertop. It has a 5.5-inch, 960 x 480 resolution display that shows things like weather forecasts, calendar events, photos and more.

These leading Natural Language Processing vendors can help AI deployments "learn" how to listen and interact with humans. As more and more companies adopt artificial intelligence (AI) in a variety of sectors, these AI are inevitably put in positions where they have to interact with human beings. From customer support chatbots to virtual assistants like Amazon's Alexa, these use cases necessitate teaching an AI how to listen, learn, and understand what humans are saying to it and how to respond. One method for teaching AI how to communicate with humans is natural language processing (NLP). Sitting at the intersection at AI, computer science, and linguistics, natural language processing's goal is to create or train a computer capable of not just understanding the literal words humans say but also the contextual implications and nuances found in their language.

Conversational Recommender Systems (CRS) has become an emerging research topic seeking to perform recommendations through interactive conversations, which generally consist of generation and recommendation modules. Prior work on CRS tends to incorporate more external and domain-specific knowledge like item reviews to enhance performance. Despite the fact that the collection and annotation of the external domain-specific information needs much human effort and degenerates the generalizability, too much extra knowledge introduces more difficulty to balance among them. Therefore, we propose to fully discover and extract internal knowledge from the context. We capture both entity-level and contextual-level representations to jointly model user preferences for the recommendation, where a time-aware attention is designed to emphasize the recently appeared items in entity-level representations. We further use the pre-trained BART to initialize the generation module to alleviate the data scarcity and enhance the context modeling. In addition to conducting experiments on a popular dataset (ReDial), we also include a multi-domain dataset (OpenDialKG) to show the effectiveness of our model. Experiments on both datasets show that our model achieves better performance on most evaluation metrics with less external knowledge and generalizes well to other domains. Additional analyses on the recommendation and generation tasks demonstrate the effectiveness of our model in different scenarios.

So let's dive into some specifics, shall we? In a nutshell, Augment adds numerous augments (get it?!) onto your existing devices. Those lowercase augments are best described as layers of intelligence that observe what you're doing and then step in as needed to make sure you always have the info you need exactly when you need it. If that concept rings a bell, congratulations: You've been paying attention. Philosophically, at least, Augment is strikingly similar to Heyday, a context-surfacing service I covered for Fast Company earlier this year.