Kartik Mohla, a graduate in management information systems, used to work at an US investment bank before he came back to India to join his family business that helps companies boost the skills and productivity of their employees.
Now, he is the CEO of InspireOne Technologies, a talent development service (spun-off from the family business) which analyses employee e-mails to let them know what sort of leadership skills they display on a daily basis.
Sankar Nagarajan, a cloud computing and analytics professional with over 20 years of experience, floated a company called Textient Analytics in late 2014 to understand how analytics and natural language processing could be used to maybe build a platform that makes sense of large sets of data.
Now, Textient Analytics is a full-fledged product that automates market research with the help of artificial intelligence programs and offers strategic insights.
Mohla, based in Gurgaon, and Nagarajan, based in Chennai, have two things in common: they built their products from scratch in under a year, and centred around IBM’s cognitive computing platform Watson. They are IBM’s first official partners for this technology in India.
International Business Machines Corp. developed Watson as a supercomputer that could compete with humans on the quiz show Jeopardy, and beat them. This required Watson to process reams of text, analyse questions posed in “natural”, everyday language and generate the correct responses.
This is termed cognitive computing, which involves computer systems that use data mining, pattern recognition and natural language processing to mimic the way the human brain works, and are ‘self-learning.’ That was in 2011.
Fast forward to 2014, and IBM announced a new business unit around Watson to push for its commercial applications. It started opening up certain application programming interfaces (APIs), computer programs which allow applications to access data from each other. Over 30 Watson services are now available on the IBM Watson Developer Cloud on Bluemix, IBM’s cloud computing platform.
One of Watson’s biggest application areas is health. IBM recently bought Truven Health Analytics, a Michigan-based provider of cloud-based healthcare data, for $2.6 billion, its fourth major health-related acquisition since launching Watson Health unit in April last year.
“With the help of Watson, organizations are harnessing the power of cognitive computing to transform industries, help professionals do their jobs better, and solve important challenges,” an IBM spokesperson said in an e-mailed reply.
“As part of IBM’s strategy to accelerate the growth of cognitive computing, Watson is open to the world, allowing a growing ecosystem of developers, students, entrepreneurs and tech enthusiasts to easily tap the most advanced and diverse cognitive computing platform available today,” the spokesperson added.
Others see more reasons for Watson tying up with start-ups.
“It’s a platform play. IBM wants to leverage newer markets like India,” said Thomas Reuner Managing Director for IT Outsourcing Research at HfS Research.
IBM Watson works with start-ups identified as “ecosystem partners” and provides them access to its APIs and support for free, and chalks out a three-year renewable, revenue sharing agreement.
It now has more than 500 ecosystem partners across 29 industries, over 100 of which have already introduced commercial cognitive enabled apps, products and services to the market.
“There are more than 77,000 developers in total who are prototyping and building cloud-based cognitive computing applications. The Watson Ecosystem is seeing tremendous global growth across a variety of industries, and India is primed for a cognitive transformation,” said the IBM spokesperson.
“I was initially looking at building a couple of things with machine learning hiring software developers and what I thought I could do is probably 10% of what we’re doing now. We really leapfrogged after embracing Watson,” says Textient’s Nagarajan.
At Textient and InspireOne, Watson powers the companies’ core products.
Usually, social media analysis around brands is limited to giving out sentiment scores or saying it was positive, neutral or negative.
Using Watson, Textient is taking consumer data and applying machine learning models trained to recognise actual human sentiment to understand how a consumer feels about a brand.
InspireOne Technologies used customer data from its parent firm to develop models to analyse e-mail communication (it is “opt-in”, which means employees have to choose to have their e-mails analysed), and let employees know what sort of leadership skills they are displaying on a day-to-day basis.
However, this sort of understanding of what can be done with data (and how) is still developing in the market.
“Just having data doesn’t help, you have to know what to do with it, and IBM has to concentrate on customer relationships and making sure this understanding is conveyed. It also has to compete, with technologies like Wipro’s Holmes, although it is much newer compared to Watson,” said Reuner.