Group items tagged

Proactive software agents will reduce the need to waste time looking for information.

Contextual search tools like Google Now, which takes into account where you are and what you are doing to provide useful information, are the first big step towards anticipatory and responsive software agents.

In the consumer world right now, Apple's Siri is the most well-known example thus far of how a software agent will interact with humans, though it has its limitations, both in speech recognition and plain common sense. As that interaction is smoothed out, though, it is not hard to imaging giving agents like Siri or Google Now's voice search more permissions to act on the information at hand, instead of just reporting it. Once that hurdle is overcome, all of that predictive and contextual information that the Internet is starting to finding for us will have a smooth, human-like interface and better able to help us manage our days.

anticipatory systems. 

Increasingly, rather than waiting for us to tell them what we want, in the form of a search query or command, they'll prompt us with suggestions.

Here's a simple definition of anticipatory systems. Think of them as artificially intelligent services that are aware of external context — including ambient inputs like time of day, social connections, upcoming meetings, local weather, traffic and more.

the goal was to introduce "conversational search." To have a conversation, you need a conversational partner.

The beauty of such Big Data applications is that they can process Web-based text, digital images, and online video. They can also glean intelligence from the exploding social media sphere, whether it consists of blogs, chat forums, Twitter trends, or Facebook commentary. Traditional market research generally involves unnatural acts, such as surveys, mall-intercept interviews, and focus groups. Big Data examines what people say about what they have done or will do. That's in addition to tracking what people are actually doing about everything from crime to weather to shopping to brands. It is only Big Data's capacity for dealing with vast quantities of real-time unstructured data that makes this possible.

Much of the data organizations are crunching is human-generated. But machine sensors — what GE people like CMO Beth Comstock called "machine whispering" when I talked with her this past summer — are creating a second tsunami of data. Digital sensors on industrial hardware like aircraft engines, electric turbines, automobiles, consumer packaged goods, and shipping crates can communicate "location, movement, vibration, temperature, humidity, and even chemical changes in the air."

the number of Google queries about housing and real estate from one quarter to the next turns out to predict more accurately what's going to happen in the housing market than any team of expert real estate forecasters. Similarly, Google search queries on flu symptoms and treatments reveal weeks in advance what flu-related volumes hospital emergency departments can expect.

Addictive technology creates “internal triggers” which cue users without the need for marketing, messaging or any other external stimuli.  It becomes a user’s own intrinsic desire. Creating internal triggers comes from mastering the “desire engine” and its four components: trigger, action, variable reward, and commitment.

A company that forms strong user habits enjoys several benefits to its bottom line. For one, this type of company creates “internal triggers” in users. That is to say, users come to the site without any external prompting. Instead of relying on expensive marketing or worrying about differentiation, habit-forming companies get users to “self trigger” by attaching their services to the users’ daily routines and emotions. A cemented habit is when users subconsciously think, “I’m bored,” and instantly Facebook comes to mind. They think, “I wonder what’s going on in the world?” and before rationale thought occurs, Twitter is the answer. The first-to-mind solution wins.

A multi-screen world, with ad-wary consumers and a lack of ROI metrics, has rendered Don Draper’s big budget brainwashing useless to all but the biggest brands. Instead, startups manufacture desire by guiding users through a series of experiences designed to create habits

online, feedback loops aren’t cutting it. Users are increasingly inundated with distractions, and companies find they need to hook users quickly if they want to stay in business. Today, companies are using more than feedback loops. They are deploying desire engines.

Desire engines go beyond reinforcing behavior; they create habits, spurring users to act on their own, without the need for expensive external stimuli like advertising. Desire engines are at the heart of many of today’s most habit-forming technologies. Social media, online games, and even good ol’ email utilize desire engines to compel us to use them.

At the heart of the desire engine is a powerful cognitive quirk described by B.F. Skinner in the 1950s, called a variable schedule of rewards. Skinner observed that lab mice responded most voraciously to random rewards.

Google Now aggregates the information Google already collects about you on a daily basis: accessing your email, your calendar, your contacts, your text messages, your location, your shopping habits, your payment history, as well as your choices in music, movies and books. It can even scan your photos and automatically identify them based on their subject, not just the file name

Google already knows where you live, for example, and constantly plots out the time it will take to return home. Google even knows your favorite routes to work and can suggest alternatives based on congestion. And it will figure out your favorite sports teams by the number of times you ask about them, without you ever having to explicitly identify them. Google’s recommendation engine, meanwhile, uses the information to suggest new content to purchase.

Google Now tries to proactively provide information via “cards,” or vertical tabs, that present information it thinks you might want. For example, if you’ve entered a home location via Google Maps, a card will constantly update with the estimated time to drive home.

Two years ago, the two greatest Jeopardy champions of all time got obliterated by a computer called Watson. It was a great victory for artificial intelligence--the system racked up more than three times the earnings of its next meat-brained competitor. For IBM’s Watson, the successor to Deep Blue, which famously defeated chess champion Gary Kasparov, becoming a Jeopardy champion was a modest proof of concept. The big challenge for Watson, and the goal for IBM, is to adapt the core question-answering technology to more significant domains, like health care. WatsonPaths, IBM’s medical-domain offshoot announced last month, is able to derive medical diagnoses from a description of symptoms. From this chain of evidence, it’s able to present an interactive visualization to doctors, who can interrogate the data, further question the evidence, and better understand the situation. It’s an essential feedback loop used by diagnosticians to help decide which information is extraneous and which is essential, thus making it possible to home in on a most-likely diagnosis. WatsonPaths scours millions of unstructured texts, like medical textbooks, dictionaries, and clinical guidelines, to develop a set of ranked hypotheses. The doctors’ feedback is added back into the brute-force information retrieval capabilities to help further train the system.

For Watson, ingesting all 2.5 million unstructured documents is the easy part. For this, it would extract references to real-world entities, like corporations and people, and start looking for relationships between them, essentially building up context around each entity. This could be connected out to open-entity databases like Freebase, to provide even more context. A journalist might orient the system’s “attention” by indicating which politicians or tax-dodging tycoons might be of most interest. Other texts, like relevant legal codes in the target jurisdiction or news reports mentioning the entities of interest, could also be ingested and parsed. Watson would then draw on its domain-adapted logic to generate evidence, like “IF corporation A is associated with offshore tax-free account B, AND the owner of corporation A is married to an executive of corporation C, THEN add a tiny bit of inference of tax evasion by corporation C.” There would be many of these types of rules, perhaps hundreds, and probably written by the journalists themselves to help the system identify meaningful and newsworthy relationships. Other rules might be garnered from common sense reasoning databases, like MIT’s ConceptNet. At the end of the day (or probably just a few seconds later), Watson would spit out 100 leads for reporters to follow. The first step would be to peer behind those leads to see the relevant evidence, rate its accuracy, and further train the algorithm. Sure, those follow-ups might still take months, but it wouldn’t be hard to beat the 15 months the ICIJ took in its investigation.