‘Kerala floods provided right context for IBM Call for Code’

There’s no better technology than artificial intelligence (AI) and machine learning (ML) for crunching the humongous data to predict weather.

Accurate predictions are pretty much possible if we get the predictive modelling right, says Seema Kumar, Country Leader, Developer Ecosystem & Start-ups, IBM India/South-Asia.

‘Can prepare better’

“We’re sitting on tonnes of data today. Maybe not all is open since as much as 80 per cent sits inside a firewall world-wide. Only the rest is available for use,” Kumar told BusinessLine here.

AI has been around existed for a long time. but why it is so relevant is because of availability of With huge volumes of data, structured or unstructured, the . We now have the technology to use it i, which is why it is important.

“We can’t stop extreme weather conditions from happening but we can definitely prepare for it better,” she said in the context of an IBM hackathon held recently for applying new technologies to deal with extreme weather.

The IBM Call for Code is a global initiative aimed at putting technology to the larger good of the mankind, and is aligned to the UN’s Sustainable Development Goals (SDGs). A theme aligned to one of the 17 SDGs is picked every year for the coding challenge.

This year it is about natural disasters.

This gains significance in the backdrop of the historic floods in Kerala in August. “This is precisely why Thiruvananthapuram was chosen to host the event. For us, coding with the aim of solving larger global issues is a way of giving back to the society,” Kumar said.

She pointed out, “If we look at the data for the last 10 years, natural disasters have had the most impact on human life, and needed to be addressed.”

The best option was to build solutions and technologies that aid disaster relief and prepare communities for similar future events as well as in rebuilding post-disaster.

People hit by the disaster, those who took part in relief operations, and the whole disaster management process had generated a lot of expert insight and ground level experience.

‘creating magic’

If coders were to internalise all this with the backup of some intelligent programming, it could create magic, said Kumar.

The Weather Company at IBM owns a comprehensive set of data. As part of the coding challenge, it had opened up this data on cloud platform to the coders.

“In fact, I don’t think there is any dearth of relevant data, which sits in various forms. All boils down to taking the right decisions using that data,” Kumar said.

Decision-makers have to derive insight from it and prioritise, remove duplication and erroneous data. AI and ML can use such insight to enable better decision-making.

“After all, AI is here to help and augment human intelligence. There’s a limit to which a human can deal with scores of data. And weather is all about data."