How cognitive computing can radically transform any industry

With Inhi Cho Suh, General Manager of Watson Work and Collaboration Solutions, IBM

Moore’s Law taught an entire generation about how fast the technology around us changes. And as artificial intelligence seamlessly integrates new technology with our workforce, the rate of change will only increase.

IBM has witnessed this rapid pace of change consistently throughout its 106-year history, and dedicates deep resources to finding ways to apply the accelerated advancements of technology to solving important world problems. The company was the first to create an automated census in the early 1900s and supplied the mainframe that supported landing men on the moon with the Apollo missions. Now, IBM Watson is focused on applying cognitive to augment human capabilities across 20 different vertical industries spanning over 45 countries.

The intersection of AI, cognitive computing and the cloud is a hot topic right now. Organizations that can harness these three technologies are poised to crack some pretty bold problems across all lines of business. Inhi Cho Suh, General Manager of Collaboration Solutions at IBM, speaks about it this way: “Fundamentally, it’s about how we can augment human capabilities by applying advanced analytics and machine learning to solve problems that impact business and society.”

Enterprises across industries are starting to use cognitive computing, AI and the cloud to mobilize data and create revelatory customer experiences that have never before been possible.

“With cognitive computing, you can radically transform entire industries. That digital disruption, combined with augmented intelligence, will change every profession and every industry.”
— Inhi Cho Suh

Grappling with dark data and messy knowledge

Cognitive computing aims to simulate human thought processes using computers that self-learn through data mining, pattern recognition and natural language processing. These systems rely on deep-learning algorithms and neural networks to understand, reason, interact and learn over time. Using progressive intelligence and a steady, abundant stream of data, cognitive computing systems get better over time. The more data, the better, if you’re trying to train an algorithm.

Most public data was written in a way that’s searchable and can be easily indexed and found. This structured data is often in HTML format, so it’s tagged and can be easily retrieved from the web. This is not the case for enterprise data. Primarily unstructured, enterprise data hasn’t been indexed in any way and is challenging to exploit — we call it dark data. Which is why, despite its abundance, enterprise data has traditionally been hard to apply to specific business use cases.

As Inhi Cho Suh puts it, “The biggest challenge my team has faced has been finding a sufficient amount of data to apply to training for particular use cases. Curating data — and curating it over time to make sure that you have the right input and output — is incredibly difficult to do.” This is changing though, due to robust cloud-computing platforms that make it possible to shine a light on dark enterprise data and marry unstructured and structured data together to create bigger, more accessible data sets.

With so much opportunity, where to start?

You have to digitize your content before you can apply analytics to it. Suddenly, disparate types of data such as text, image and video files can be combined in the cloud. Leveraging cloud content management technologies, cognitive computing projects are able to combine different types of data such as transactional data (what goes into your financial ledgers) and cognitive behavioral data (like the activities of your sales associates in your storefront).

Being able to harness diverse structured and unstructured data in the cloud like this creates a whole new opportunity for enterprises. Now, teams like Watson can embark upon cognitive computing projects across industries, from the financial services sector (around credit and risk), to the retail sector (pricing and customer engagement) and to healthcare (quality of care, diagnosis and recovery).

Cognitive computing is becoming a democratic technology that promises to solve some of the greatest pressures C-suite and IT leaders are under today, particularly the need to differentiate their organizations and provide customers with better solutions quicker. “Frankly,” Suh says, “You just have to pick a use case to get started.” One example is the work IBM Watson team has done with H&R Block.

Tangible use cases for cognitive

Tax planning and investment counseling are stressful endeavors — not just for the taxpayer, but also for the preparer, who is under pressure to evaluate every factor in the equation and make the best decision for each unique scenario. There are a lot of inputs, and no one formula that will work for everyone because individual people have unique financial priorities. One person might be focused on saving for a child’s college fund, while another is invested in an elderly parent’s end-of-life care. Financial situations vary, and laws change constantly from state to state in the U.S. All of these variable and fluctuating factors contribute to financial and tax decisions.

Companies like H&R Block are using cognitive computing to provide a next level of innovation and dramatically enhance the work its already doing for customers in the retail office experience. Cognitive computing augments human capabilities, providing a tool that can elevate the decision-making of people and boost the vast creative capacity we have as humans. By inserting Watson’s algorithms into the relationship with taxpayers and preparers, H&R Block is better able to ensure optimal outcomes.

“One of the biggest surprises in this space [of Cognitive] is the sheer creativity, breadth and energy that everyone is bringing to it. It’s really about augmenting human capability.”
— Inhi Cho Suh

Leaders from all walks driving innovation

During these early stages of applying cognitive to business use cases, a number of roles are emerging. The CIO or CTO is an obvious leader in this arena. “But often,” Suh confides, “There are other experts, like the head risk and credit leader for a financial institution, who are looking at a number of financial elements in terms of regulatory compliance, or a head of product who partners very closely with the head of technology, or customer service leader.”

Regardless of what industry or organization they’re in, leaders are feeling the pressures of speed — speed of competition, speed of innovation and even the speed of their own internal agility to develop products and services more quickly. These pressures are creating a need for a next level of creative innovation — tools to help them augment the work they’re already doing. They need technological capabilities that are reflective of where their industries are headed in terms of things like micro-services — agile services you can easily compose.

For leaders across organizations, data is the “the new differentiator” in business — the factor that will make companies stand apart.

How to get started with cognitive

Now that cloud platforms have democratized accessibility to enterprise data, there aren’t a lot of limitations to what companies can envision. IBM Watson already has a diverse roster of cognitive computing projects in the works that touch on fields from music production to healthcare.

The first step to embarking on any cognitive computing project is to understand where your valuable differentiated data exists. Suh says, “In my view, it’s in your people, meaning, the expertise they have. It’s in your engagement platforms and the content people generate.” Both inter- and intra-company interactions between people create this valuable data.

Depending on the nature of your data sets, you then need to find a platform to curate your data and apply analytics too. Analytics might mean text analytics, natural language processing, or speech, image or voice recognition, for example. The spectrum is large, but again, choosing your most viable use case is the starting point.

“Data, methodology, governance, and end-user experience — all four are incredibly difficult to do, but once you get all four right, I mean, the outcome is amazing.”
— Inhi Cho Suh

The workplace of the future

We’ve stepped into a time when the pressures of speed, innovation and competition are enormous. In business, cognitive computing’s role is clear: to help unleash a whole different side of the creative human capacity, help us interact at a deeper level and extract meaningful insights so that every conversation with customers counts. Being able to quickly assemble services you need in the cloud and take advantage of neural networks architected for AI and cognitive computing give today’s enterprises across industries an advantage.

Cognitive computing in the cloud has the potential to help businesses across all industries create better end-user experiences and ultimately solve some pretty audacious goals around healthcare, education, IoT and other enormous societal challenges.

“I see cognitive computing everywhere in the workplace of the future. There isn’t a part of the way we work that won’t be instantiated in some form of code.”
— Inhi Cho Suh

The blueprint for cognitive computing implementation

1. Understand where your differentiated data lives. For most companies, Suh says, “It’s in your people — the expertise they have and the content they produce.”
2. Know exactly what problem you want to solve for your end users (but be willing to pivot down the line if your first hypothesis doesn’t hold up).
3. Have the right content governance in place in terms of expertise, methodology and technology. Find the best cloud-based platform to enable curation of your differentiated sets of data, and build your analytics on top of that.

About Inhi Cho Suh

Inhi Cho Suh is the General Manager of Watson Work and Collaboration Solutions at IBM, where she works to infuse analytics across workplace applications. Prior to that, she was the Vice President of Analytics Strategy and Business Development at IBM, among other roles. She holds a J.D. from North Carolina Central University School of Law.