Google Changes Search, Moves Closer To Becoming Artificial Intelligence Engine

Google began rolling out a feature that gives searchers in the United States the potential to access more relevant and in-depth responses to answers without leaving the page. The concept is built on something the company calls "knowledge graph," which ties together words to create relationships.

There are a multitude of sources behind this data. The search results page displays a variety of content related to keyword queries, bringing up a list of facts, photos, and landmarks, as well as quick links to other popular uses for the search term. Think of a Web beneath the user interface layer of the Internet that ties together all information across the Web.

Rob Garner, vice president of strategy at agency iCrossing, said Google's knowledge graph takes another step in the company's long transition to develop an artificial intelligence engine -- semantic search. "It's something Google's doing in parallel to Schema.org in terms of relating object, places and people," he said. "Looking at the schema for a person you can actually define the relationship with other people using schema vocabulary."

For example, someone looking for information on Marie Curie will see her birth and death dates, but also details on her education and scientific discoveries. The search engine understands much more about Frank Lloyd Wright than the word connecting the characters. It understands that Wright was an architect who was born June 8, 1867 in Richland Center, Wisconsin, and died on April 9, 1959, in Phoenix. It knows that his projects included Robie House and Taliesin West.

The change represents an effort by search engines to move away from text-based links in search results and serve up knowledge in fewer clicks. Last week, Microsoft rolled out a revamped version of Bing that will initiate new ad formats.

Garner said it's not clear what the advertising impact will be. "It will be interesting to see if there's a connection to better networking information around companies and product offerings, but they do that pretty well in standard Web results," he said.

Source : mediapost

Read more »

How artificial intelligence could create a personalised and perfect travel service

Feels a little creepy – doesn’t it? But, this is the future of air travel (and how we will interact with technology in general).

Airlines will know customer preferences, what “home” means for a traveler, if he likes aisle seats on morning flights, and – most importantly for use on third-party aggregators – what his preferred airline is.

The kicker: these conversations won’t be happening with live representatives drawing from a knowledge base. Passengers will instead converse with avatars, equipped with artificial intelligence and natural language understanding.

Recent news of life-sized holograms at three major New York airports has many passengers considering artificial intelligence and the technology’s real implications on their travel experience.

Minds go to fanciful extremes like the Terminator or Minority Report and they wonder: “What do these things do?” or “What’s coming next?”

Well to begin, the NY installations won’t be having true conversations with anyone.

And they also aren’t the first of their kind in the travel industry: Manchester Airport became the first to use holograms back in 2011 for security preparation.

United (via Alex) and Alaska Airlines (via Jenn) use enhanced conversational avatars online, to interact with customers, answer questions, and assist with the booking process.

Conversational avatars are already engaging with customers everyday and present the path towards (counter intuitively enough) personalized service.

The massive amounts of data that can be gleaned and analyzed from logged conversations give a true window into what customers really want.

Rather than operating within the confines of links and search terms – customers can simply ask for what they’re looking for. With that, airlines begin to measure customer desire and store preferences.

Transforming customer information into personalized service is a key factor in the future of customer loyalty.

Why do it?

Currently, passengers view flight as a commodity, where price and schedule are more important than brand. But beyond this hunt for the lowest fare, customers are searching for personalized service.

They want their individual needs to be recognized and will support a brand that demonstrates that itunderstands their expectations, needs, quirks, and specific requests.

At the same time, sensitivities around privacy and storing personal information, such as what “home” means, makes the integration a balancing act.

In order for artificial intelligence to be truly effective in building customer loyalty in the future, airlines must be sure that transparency is built in around the technology.

Building a strong customer relationship is contingent on the amount of passenger information an airline can store, understand, and act upon – but passengers want to know what information is being stored and have control over how their preferences are determined.

There is no other way to deliver this transparency other than showing a customer the process behind an avatar’s decision making process.

This is key to successful integration – rather than take control away from passengers, make the planning and booking process more transparent by engaging in conversation and revealing how actions are taken on the passenger’s behalf.

This is part of the foundation to making artificial intelligence effective; it’s not just about the technology driving it, success relies on the trust that customer’s must develop on their own.

The Manchester and NY holograms barely scratch the surface of the use of artificial intelligence in the travel industry.

The real benefits come when the conversations between customer and technology begin.

Airlines that interact and build relationships with passengers through customer experience technology will lead the brightest future in winning customer loyalty and the lifetime revenue that comes with it.

Source : This is a guest article by Dr Charles Wooters, chief scientist of Next IT.

Read more »

The brain and the Memory- Prediction framework theory

The full memory-prediction framework theory was first introduced by Jeff Hawkins in the book On Intelligence. The theory says that the physical arrangement of brain cortex tissue is uniform and means that there is a single principle that defines all brain and memory processing. It also notes that the brain's intelligence comes from the ability to predict future events by relying on past data.

Memory-prediction framework theoryThe memory-prediction framework gives a unified theory about complex behaviors and allows us to better understand what intelligence is.

This theory focuses on the cortex because we're only concerned about what makes us intelligent and not how our entire body works. After all we're trying to create intelligent software, not recreate human beings.

The Central Concept

Memory-prediction framework posits that inputs coming from the bottom of the hierarchy to the top are analyzed and ranked in a hierarchy of recognition. This then invokes a list of expectations ranked in order of potential. The framework comes into play when the brain has to compare and match up these inputs and expectations. The memory-prediction framework means that the brain does not have to consider every option at every level of the process because it uses past sequences as a guide to predict likely future sequences.

"Memory-prediction framework posits that inputs coming from the bottom of the hierarchy to the top are analyzed and ranked in a hierarchy of recognition."

The further up the framework, the longer the past sequences are and consequently the fewer options there are to finish them so the process actually accelerates as it nears the end. For example in looking at a scene, the brain first recognizes lines, then shapes and colors and finally recognizes them as objects. At the same time in the same framework, predictions about what to expect from these objects flows down to speed up our interpretation of the scene.

The framework changes as we age because we add new memories to the system. Since we start with none, as babies we truly see things for the first time. But as we age, we collect memories and this bank of expectations about the way things are and the way things work. This helps us to understand the world around us and to process external information and stimuli more rapidly, but it helps explains how different people can interpret the same situation differently because they bring different memory structures to the experience.

The Pioneer of the Memory-prediction framework theory

Jeff Hawkins originally trained as an electrical engineer and perhaps this gives us a bit of insight on how he approached the problems of brain theory and discovered the memory prediction framework theory. His theory is one of the brain as an organ capable of future predicting and error correction. The brain predicts future events by relying on past data.

"His theory is one of the brain as an organ capable of future predicting and error correction."

The system is a hierarchy so the steps of the analysis are performed in order and if the current even deviates from past experience at any level then a new string of events, or memory, is created. This string can then be used in the analysis of future situations. In this way, the system itself learns and evolves.

It continuously grows more complex and better at making predictions. Just as the theory describes a brain that is constantly adding details and sequences of understanding, the theory itself needs to grow and be fleshed out with details before it will be fully accepted.

Read more »