Is netflix the evolution of cognitive search?
Every movie recommended to you on the Netflix streaming service comes from an algorithm tuned via machine learning. And it turns out that of all the money being spent on AI research globally--$26 to $39 billion in 2016 according to McKinsey--machine learning received the biggest investment.
Machine Learning Makes Cognitive Search Smarter
Apparently, Netflix users have short attention spans when it comes to searching for videos--60 to 90 seconds. So, Netflix prefers to hook them with recommendations based on past viewing.
User behavior is much more predictive of viewing preference than user ratings. In other words, what people say they like and want to watch is different from what they actually watch. This realization has saved the company an estimated one billion annually in reducing canceled subscriptions.
In the enterprise scenario, attention span isn't so much of an issue, but productivity is. Time spent searching is time not spent translating insight into action--so past search behavior is a mother lode of valuable data for search applications just like past viewing behavior is for Netflix.
Context... the X Factor
While discussing the Netflix recommendation engine, a 2013 Wired magazine article also explored the issue of context. Netflix had noticed that viewing behavior could differ by day of the week, time of day, device, and location.
Think about that in terms of cognitive search. If you're looking for information about same store sales on Monday, it could be for a completely different reason than if you were looking on Friday. On Monday, maybe you need general trends for an all-hands meeting. But on Friday, you need very detailed statistics for a report. Or, on Monday, you're almost always in the office. And, on Friday, you're usually on the road.
Eventually, the algorithms should "learn" this and adjust your search results accordingly.
But, context is "squishy." And, Netflix acknowledged there were "practical challenges" to coding for that.
Netflix for Knowledge Management?
A recent blog on TechCrunch suggested that what we really need is a Netflix for knowledge management (KM). And there's no conceptual reason why the Netflix recommendation model couldn't be applied to KM.
The blog quotes former GE CEO Jack Welch, who said, "An organization's ability to learn, and translate that learning into action rapidly, is the ultimate competitive advantage." Increasing that ability would be the focus of a Netflix for KM. And machine learning would play a significant role.
It's likely that several AI technologies would play a role in this vision of KM and it would have to grow beyond the Netflix model.
Adaptive Learning, Machine Learning, and Cognitive Search
When employees search for information to do their jobs, they're not in a "class," but they are learning. They're trying to complete a task--an "assignment." Cognitive search, aided by machine learning and natural language processing, tailors search results based on an individual's past search behavior. Just like Netflix.
Looked at more broadly, cognitive search contains some of the features of adaptive learning. McGraw Hill describes adaptive learning like this:
Imagine that you could give every learner their own personalized course, made specifically for their strengths, weaknesses, goals, and engagement patterns. Imagine a course that adapted in real-time to their activity and adjusted moment by moment to their performance and interest level.
Or, more simply
Adaptive learning is an education technology that can respond to a student's interactions in real-time by automatically providing the student with individual support.
It's easy to see how the next generation of cognitive search solutions could deliver this kind of an experience to employees in the context of their jobs. The cognitive search application becomes a virtual tutor. The Netflix analogue would be a movie avatar that interacts with a customer one-on-one. Underneath both--machine learning and natural language processing.
The Danger of Over Promising
Just as in the early days of Big Data, the hype surrounding AI and its supporting technologies is enormous. A good example would be IBM's Watson. The glow of its Jeopardy! and chess wins has certainly faded. Recently, it was the focus of a very negative report by global investment banking firm Jefferies.
It's not that there's no technology behind Watson; it's that what has been promised has often not been delivered. Or, at least, not at a cost that made the investment worthwhile.
Back in 2016, Bernard Marr writing for the #BigData tech blog on Forbes noted that machine learning is the "field of AI which today is showing the most promise at providing tools that industry and society can use to drive change."
That's still true today. Right now, machine learning and its close cousin deep learning are the AI technologies making the most impact in business. Many businesses are investing in machine learning as first steps toward a fuller embrace of AI.
As McKinsey points out, there's typically a gap between investment in new technologies and their commercial application. The current state of AI reflects that gap.
McKinsey segments the AI market by adopters, partial adopters, contemplators and experimenters, with the largest segment being contemplators at 40 percent. Regardless of where your organization falls, it's safe to say that to remain competitive AI will need to become part your strategic plan.
Attivio is the leading Cognitive Search and Insight Platform company. Our Fortune 500 clients rely on us to drive innovation, operational efficiencies, and improve business outcomes.
By Lou Jordano, CMO, Attivio
|Printer friendly Cite/link Email Feedback|
|Date:||Sep 1, 2017|
|Previous Article:||Cognitive computing: film at eleven.|
|Next Article:||How to steer cognitive applications with semantic technologies.|