Traditional recommendation systems are faced with two long-standing obstacles, namely, data sparsity and cold-start problems, which promote the emergence and development of Cross-Domain Recommendation (CDR). The core idea of CDR is to leverage information collected from other domains to alleviate the two problems in one domain. Over the last decade, many efforts have been engaged for cross-domain recommendation. Recently, with the development of deep learning and neural networks, a large number of methods have emerged. However, there is a limited number of systematic surveys on CDR, especially regarding the latest proposed methods as well as the recommendation scenarios and recommendation tasks they address. In this survey paper, we first proposed a two-level taxonomy of cross-domain recommendation which classifies different recommendation scenarios and recommendation tasks. We then introduce and summarize existing cross-domain recommendation approaches under different recommendation scenarios in a structured manner. We also organize datasets commonly used. We conclude this survey by providing several potential research directions about this field.

In this paper, we propose an end-to-end sentiment-aware conversational agent based on two models: a reply sentiment prediction model, which leverages the context of the dialogue to predict an appropriate sentiment for the agent to express in its reply; and a text generation model, which is conditioned on the predicted sentiment and the context of the dialogue, to produce a reply that is both context and sentiment appropriate. Additionally, we propose to use a sentiment classification model to evaluate the sentiment expressed by the agent during the development of the model. This allows us to evaluate the agent in an automatic way. Both automatic and human evaluation results show that explicitly guiding the text generation model with a pre-defined set of sentences leads to clear improvements, both regarding the expressed sentiment and the quality of the generated text.

Fox News Flash top entertainment and celebrity headlines are here. Check out what clicked this week in entertainment. Kristin Cavallari is going for the online dating route following her split from Jay Cutler. "People have set me up. Mutual friends," she said on the "Not Skinny, But Not Fat" podcast episode on Tuesday.

Amazon took up a lot of oxygen in the deals room last week with Prime Day. However, there are a few gadgets and gizmos that are even cheaper this week than they were then. If you're in the market for a new robot vacuum, keyboard, or even a large TV, we've got some goodies for you. Special offer for Gear readers: Get a one-year subscription to WIRED for $5 ($25 off). This includes unlimited access to WIRED.com and our print magazine (if you'd like). Subscriptions help fund the work we do every day.

A version of this post was originally published in Entrepreneur on February 1, 2022. I've been in the education business for decades as a senior lecturer, trainer and CEO. When people ask me about the biggest challenge that learners face, the first thing that comes to mind is that learners see training as something they "have to do." Now, let's think for a moment about this. How did we get here? Why aren't we talking about "want to do" or "happy to have the opportunity to do?"

Nowadays, recommender systems and search engines play an integral role in fashion e-commerce. Still, many challenges lie ahead, and this study tries to tackle some. This article first suggests a content-based fashion recommender system that uses a parallel neural network to take a single fashion item shop image as input and make in-shop recommendations by listing similar items available in the store. Next, the same structure is enhanced to personalize the results based on user preferences. This work then introduces a background augmentation technique that makes the system more robust to out-of-domain queries, enabling it to make street-to-shop recommendations using only a training set of catalog shop images. Moreover, the last contribution of this paper is a new evaluation metric for recommendation tasks called objective-guided human score. This method is an entirely customizable framework that produces interpretable, comparable scores from subjective evaluations of human scorers.

Amazon's Alexa voice assistant will soon tell users when items they've had an eye on are about to be on sale. The company announced on Thursday that Alexa will notify Prime customers up to 24 hours in advance of upcoming sales on items in their Amazon wish list and shopping cart, or items that were marked "saved for later" on the platform. The feature will work with fourth-generation Echo smart speakers and newer. For example, if a customer is looking at a new TV and it's been sitting in a wish list, Alexa will give the consumer a heads-up that it will be on sale soon. The Amazon Echo ring will turn yellow when an item goes on sale, so a user knows to ask Alexa to read out their notifications.

This week brought back some of our favorite tech deals we've seen all year. A four pack of Apple's AirTags is back on sale for $89, while the Apple TV 4K has dropped to $130. While neither of those are all-time lows, they're very close and great deals on a couple of the most popular Apple gadgets right now. Amazon also just discounted its Echo smart speaker to $60, which its a record low and a return to its Prime Day price, plus you can still pick up the Chromecast with Google TV for only $40. Here are the best tech deals from this week that you can still get today.

If you missed the chance to pick up an Echo smart speaker during Prime Day last week, you have another opportunity to do so today. Amazon's full-sized Echo is back down to a record-low price of $60, which is 40 percent off its normal rate. The Echo Show 5 is also on sale for $40 right now, which is only $5 more than it was on Prime Day. The Echo Dot, on the other hand, is currently 20 percent off and on sale for $40. You may just think of Amazon's Echo as a way to get Alexa into your home, but it's a pretty capable speaker as well.

While rich, open-domain textual data are generally available and may include interesting phenomena (humor, sarcasm, empathy, etc.) most are designed for language processing tasks, and are usually in a non-conversational format. In this work, we take a step towards automatically generating conversational data using Generative Conversational Networks, aiming to benefit from the breadth of available language and knowledge data, and train open domain social conversational agents. We evaluate our approach on conversations with and without knowledge on the Topical Chat dataset using automatic metrics and human evaluators. Our results show that for conversations without knowledge grounding, GCN can generalize from the seed data, producing novel conversations that are less relevant but more engaging and for knowledge-grounded conversations, it can produce more knowledge-focused, fluent, and engaging conversations. Specifically, we show that for open-domain conversations with 10\% of seed data, our approach performs close to the baseline that uses 100% of the data, while for knowledge-grounded conversations, it achieves the same using only 1% of the data, on human ratings of engagingness, fluency, and relevance.