Sign up for our newsletter and get the latest big data news and analysis.

Machine Learning: Better Means of Consumption Required

Finding critical insights in today’s vast and varied pools of data is akin to finding a needle in a haystack.  While business users search for the elusive needle that is insight, they are missing out on opportunities to drive revenue and grow the business. This is why, increasingly, companies turn to machines to locate the “needles” (thank you Mythbusters for proving modern technology can in fact help sort through the hay much faster). With the “finding” part covered, business users can focus on what do with the needles rather than searching for them.

IDC recently forecast worldwide revenues for cognitive and artificial intelligence systems will reach $12.5 billion in 2017 and more than $46 billion in 2020.  The largest area of spending in 2017 will be on cognitive applications – which include machine learning – at $4.5B. Yet, despite embracing machine learning, many businesses haven’t figured out how to use it effectively.

One key challenge in successfully using machine learning is figuring out how to make it accessible to end users. For it to have an impact on the business, machine learning needs to become more consumable. It needs to be exposed, guided learning that starts with surfacing data for user validation. This is followed by engaging users in determining if the data makes sense given the situation. These two steps not only guide (and drive buy-in from) users, but also help the machine(s) learn.

Companies that will win using machine learning will be those that figure out how to take the innovation, apply it in simple ways, and make it consumable for the end user.  It’s not just the best algorithm that provides a strategic advantage, it’s the creation of an experience that makes the information consumable and actionable by the largest audience.

Think about a self-driving car: it’s not the technology, but the experience that stimulates adoption. If the car is difficult to use, or drivers don’t feel comfortable behind the wheel, they’re less likely to trust the car.  If the experience is user-friendly and inspires confidence, drivers will more readily give the concept a chance. Today’s cars are slowly exposing drivers to the technology through specific use cases such as automatic emergency braking, where the car stops itself when it senses a collision.

The willingness to adopt machine learning in business will also initially center on specific use cases.  Some use cases have very small margins of error so will require confidence in both the experience and the results before adoption. Other use cases can more readily absorb the errors associated with “learning on the job” and, thus, will be more logical starting points for introducing machine learning to the enterprise.

An example of the second use case, in Financial Services, early adoption will likely center on supporting processes such as reconciliations, spotting outliers or anomalies in the data, and recommending adjustments; forecasting and planning; and integrating data across sources to identify forward-looking indicators, which enable the office of finance to see the early warning signs of impending trouble.

Other industries, such as retail, will use machine learning to aid in customer journeys and individualizing the customer experience based on the vast amounts of data collected online, in stores and through IoT to increase sales and engender loyalty.

We live in a world of adaptability and experience (thank you, Steve Jobs), and our expectations for technology have well surpassed simply answering, “does it work”. Technology must be easily consumed, intuitive and not make us think too hard. After all, finding that needle is such a challenge because it’s difficult. Why would vendors NOT focus on the experience of making the task easy as a core tenet of their design process?

About the Author

Darren Peirce is CTO of Magnitude Software. Darren drives Magnitude Software’s technology direction and product delivery. With responsibility for identifying the nexus of market opportunity, technology innovation and Magnitude’s current and future portfolio, Darren oversees Product Management, Product Delivery and Engineering Services while developing and maturing key partnerships with leading technology companies in support of product strategy. He received a degree in Management from Intec College in Cape Town, South Africa, and studied Economics at the University of Stellenbosch.

 

 

 

Leave a Comment

*

Resource Links: