The volumes of data generated every day won't reveal anything in and of themselves. Human expertise is needed to make sense of them. Industries across the world are more connected than ever. Companies that operate internationally are reliant on a wider sphere of influencers, providers and partners. All of these relationships create the potential for business insight, but how do you tap into that data?

In this work we propose a data-driven modelization approach for the management of advertising investments of a firm. First, we propose an application of dynamic linear models to the prediction of an economic variable, such as global sales, which can use information from the environment and the investment levels of the company in different channels. After we build a robust and precise model, we propose a metric of risk, which can help the firm to manage their advertisement plans, thus leading to a robust, risk-aware optimization of their revenue. The advertising industry represents an estimate of US$ 529.43 billion [4] and this quantity is likely to increase in the following years. In parallel, there has been a recent interest in the application of data-driven models in the context of forecasting and decision making in the marketing industry.

Recurrent neural networks are nowadays successfully used in an abundance of applications, going from text, speech and image processing to recommender systems. Backpropagation through time is the algorithm that is commonly used to train these networks on specific tasks. Many deep learning frameworks have their own implementation of training and sampling procedures for recurrent neural networks, while there are in fact multiple other possibilities to choose from and other parameters to tune. In existing literature this is very often overlooked or ignored. In this paper we therefore give an overview of possible training and sampling schemes for character-level recurrent neural networks to solve the task of predicting the next token in a given sequence. We test these different schemes on a variety of datasets, neural network architectures and parameter settings, and formulate a number of take-home recommendations. The choice of training and sampling scheme turns out to be subject to a number of trade-offs, such as training stability, sampling time, model performance and implementation effort, but is largely independent of the data. Perhaps the most surprising result is that transferring hidden states for correctly initializing the model on subsequences often leads to unstable training behavior depending on the dataset.

The digitally driven retail environment has caused lots of adapt-or-die challenges for payment processors and merchant acquirers, including spotting fraud as it moves from one device and channel to another. "One of the attacks that I've seen in particular is to perform a low-value fraudulent transaction in one channel to build trust from within the system, then do a larger-value transaction elsewhere," said David Excell, founder of Featurespace, a British company that feeds artificial intelligence into behavior analysis in an attempt to track, predict and stamp out fraud that moves among channels. "There are still friction points that should be solved," Excell said. "Just last week my credit card was blocked because of a recurring monthly insurance payment that had been on the card for four months. That's where you can apply AI. To improve spotting that and make the experience better. But there's still a long way to go." Find out how ARIC Fraud Hub fights fraud with adaptive behavioural analytics & book a demo.

Apple's delayed $349 'HomePod' smart speaker is finally set to go on sale within weeks, it has been claimed. The gadget will battle Amazon's Echo and Google Home for the lucrative smart speaker market, using Apple music and Siri to do everything from play music to give news and traffic updates - but was delayed from its inital launch date of November. The firm said it needed'a little more time before it's ready' - and now, a leading analyst says the firm is ready. The new $349 smart'HomePod' home speaker will go on sale later this year, and use Siri to aplay music and answer questions. Apple also unveiled iOS 11 and new iPads at the event.

Samsung Electronics just announced its new artificial intelligence technology for its high-end QLED TV series. The technology has advanced machine learning capabilities that make it possible for the South Korean giant's TVs to render any type of video content with better quality. Ahead of Consumer Electronics Show (CES) 2018, Samsung unveiled its new 164-inch "The Wall" TV at the Enclave event venue in Las Vegas on Sunday. Interestingly, the company also showcased other innovations and products at the event including the world's first 8K AI technology, as indicated by Samsung in its press release. Samsung's new invention is its response to the rapidly growing number of super-high resolution television screens in the market in spite of the scarcity of readily available high resolution content. With Samsung's 8K AI technology, any type of video can easily be upscaled and transformed into 8K content.

At the end of last year Hisense launched the $10,000 100-inch 4K "Laser TV" projector we saw at CES 2017, and this year it's continuing to push that home theater in a box. According to president Liu Hong Xin, it has more sizes in development (80 and 88-inch sizes will be in its booth), and already owns 60 percent of the market share for 85-inches and above. As far as regular TVs go, the company once again has a wide range of LED-lit LCD televisions ranging in size and price, however the biggest change is that there's no mention of Roku built-in. This time around the high-end H10E (above, which features over 1,000 zones of LED lighting) and H9E Plus series will feature Android TV, while its other models come with the Hisense TV Platform. It's not clear what Hisense is using for that, but it will have apps for Netflix, Amazon, Pandora, YouTube and Fox Sports Go. All of them are Amazon Alexa enabled with voice control access, while the Android models also support Google Assistant.

CES was a different show 17 years ago. In 2001, the Best of Show award went to the DataPlay disc, a postage-stamp-size memory card that held up to 500MB of data. If you can't remember it, don't worry -- the company went out of business shortly thereafter as SD became the de facto standard. Microsoft was still the biggest tech company on the planet at that point -- Apple wouldn't release the original iPod for another nine months -- and each CES began with a keynote hosted by Bill Gates. The first 80 minutes of its 2001 keynote consisted of clunky tech that now, in some form or another, lives in Google Home or your smartphone.

This could create security nightmares. Worse still, it could strip away from each of us a part of our uniqueness. But companies, universities and governments are already working furiously to decode the human voice for many applications. Technologically adept nation states (the US, China and Estonia) have waded into this space and tech giants such as Google, Amazon, Apple and Facebook also have special projects on voice. It's not that hard to develop an artificial voice, then model and reproduce spoken words and phrases.

Pop artist Taryn Southern, who appeared on American Idol in 2004, created the lyrics and melodies for "I AM AI" but left most of the other work to software programs. The album's first song "Break Free," which was released on Monday, was developed with the help of startup Amper Music. Amper is one of several AI music services Southern is working with on the album, which will debut later this year. Southern has only basic piano skills, so she turned to the program to deliver the instrumental part of the song. The AI developed the harmonies, chords and sequences.