For example, AI and automation allow insurers to cut down on claim processing and underwriting times significantly and reap sizable cost savings. Tasks that once took months to finish are now accurately completed in the matter of minutes, opening the gate for insurers to focus on more complex and creative projects. Machine learning can help insurers and agents underwrite risk more effectively, using the large troves of customer data it has collected.

The robots that think: China's most advanced assistants go on show Artificially intelligent robots can now understand the way humans think. Pictured above, a child shouts at the companion robot'Canbot,' also on display at the World Robot Conference Above, a woman demonstrates the ability of Baxter, an industrial robot from U.S. company Rethink Robotics, to follow her hand movements Dubbed the'robot goddess', Jia Jia is being taught deep learning abilities, including understanding human language, and detecting facial expressions. When asked by an audience-member'What kind of skills do you have?', Jia Jia replied: 'I can talk with you. I can identify gender and age of people standing front of me, and I can detect your facial expressions' The humanoid robot is programmed to recognise human and machine interaction, with autonomous position navigation and services based on cloud technology.

Robots, self-driving cars, drones, bots and many IoT devices are becoming smarter with Machine Learning. She brings twenty years of digital transformation experience from building organizations, shaping new technology ecosystems and mentoring leaders at eBay, PayPal, Harcourt, and GTE. She is an advisor for Barcelona Technology School and select startups. She has an MBA from Boston University. Register for @[email protected] 'FREE' Before Friday!

When it comes to enterprise customer care, machine learning enables virtual assistant solutions to automate tasks that used to require a live agent: password resets, address and complex information collection, even sales support. Integrating machine learning into enterprise customer care opens doors to more flexible automated solutions. With the growing challenges and volume of customer interactions that most companies must handle, that flexibility, efficiency, and accuracy is exactly what's needed. By incorporating natural language processing, many of today's automated customer care solutions can more accurately understand what callers are saying.

The newest generation of machine learning-enabled applications is transforming how companies are using technology to get closer to customers, giving people more time for innovation so they can create even more valuable products and services. Consider Internet cafes that fill seats with streamlined ease using chat bots as digital concierges. Or supermarket shelves that let sales reps at consumer packaged goods (CPG) companies see exactly what's selling (or not) in real time -- no physical sensors needed. These are just a couple of the exciting machine learning demo spotlights in my video interview interview at SAP TechEd Las Vegas with Karsten Schmidt and James Rapp, both of the SAP Innovation Center in Silicon Valley. As I listened to Schmidt describe how customers, developers and others are interacting with HanaHaus 24/7 using chat bots to reserve space more efficiently, I was struck by how much this innovation redefines community.

Mobile apps like Facebook, Twitter, Snapchat and Instagram are touted as the next great opportunity for advertisers, publishers and businesses of all types. But what if a simple conversation tool had the ability to turn prospects into customers far better than any of those hot platforms? Chatbots are software programs that use messaging platforms as the interface to perform a wide variety of tasks--everything from scheduling a meeting to reporting the weather, to helping a customer buy a sweater. Because texting is the heart of the mobile experience for smartphone users, chatbots are a natural way to turn something users are very familiar with into a rewarding service or marketing opportunity. And when you consider that the top 4 messaging apps reach over 3 billion global users (MORE than the top 4 social networks), you can see that the opportunity is huge.

A few weeks ago I posted a map of the Artificial Intelligence landscape in Ireland. The map is a visual representation of resident Irish companies in the A.I. space. I decided to go a step further and put together an overview of the Irish Artificial Intelligence industry. I used the companies on the map to form the basis of the findings covered in this blog. I use the phrase'Artificial Intelligence' as an umbrella term to categorise the industry.

One approach to building conversational (dialog) chatbots is to use an unsupervised sequence-to-sequence recurrent neural network (seq2seq RNN) deep learning framework. About a year ago, researchers (Vinyals-Le) at Google published an ICML paper "A Neural Conversational Model" that describes one such framework; a review can be found here. The Vinyals-Le paper (and associated framework) is instructive in understanding some of the parameters of such seq2seq chatbot models. We assume that Vinyals-Le used Tensorflow, though this is not explicitly stated in the paper. Note that seq2seq may not be the best way to build a truly conversational chatbot; the Vinyals-Le chatbot is more of a Q/A system that originated in machine translation.

Last week, we talked about some changes we're bringing to our blog. Our goal is to create a forum where Artificial Intelligence (AI) can be examined and explored fairly without being oversimplified or overhyped. With Natural Language Generation (NLG), one of the most common storylines we see discussed in the media is how writing jobs will be impacted. Some argue that AI software will replace creative writers and journalists. This could not be further from the truth.

A group of researchers from the University College London (UCL), University of Sheffield, and University of Pennsylvania, created an Artificial Intelligence system to judge 584 human rights cases and had released its findings recently. The cases analyzed by the AI method were previously heard at the European Court of Human Rights (ECHR) and were equally divided into violation and non-violation cases to prevent bias. European Court of Human Rights in Strasbourg, France (Image Credit: ECHR) So how did the AI judge perform? Basing its judgment on the case text, the AI judge managed to predict the decisions on the cases with 79% accuracy. That means, it concurred with the decisions of human judges 8 out of 10 times.