Open data essential for machine learning

By Lee Min-hyung

The government-led open data initiative is essential for helping local companies gain a competitive edge in the global machine learning industry, market insiders said Thursday.

Machine learning is one of the most promising technology fields identified as the next driving force by global technology giants such as Google, Microsoft and Facebook. As a type of artificial intelligence, the technology uses massive datasets to make humanlike predictions.

But the machine learning industry here has yet to meet the global standard even in the presence of the world’s fastest mobile data service infrastructure.

“A lack of massive amounts of big data is the biggest stumbling block to prevent local tech companies from leaping forward in the area,” said Eom Jae-hong, manager at SK Telecom’s Artificial Intelligence (AI) tech lab, Thursday.

“We have limited access to big data, as the government is reluctant to share it with private sectors,” he said. “In the case of the United States, the government is a strong supporter of IT giants, helping them develop software using the government’s big data.”

Google can collect immense amounts of data from around the world with its global search engine and Gmail service, enabling the company to lay down infrastructure for machine learning.

“But it’s very tough to collect big data here due to government regulations over sharing of information,” he said.

In Korea, only a few companies, such as Naver and SK Telecom, are investing in the machine learning industry.

“Korean companies have neglected machine learning, as it was of no use in terms of profitability,” said another source, asking not to be named.

“Companies here cannot expand investments in the area, as there are no large enough sets of data or infrastructure to manage and analyze datasets, which means they cannot create a business model using machine learning.”

Naver, the nation’s largest portal service operator, is the leader in the local machine learning industry. Naver Labs, the company’s research and development arm focusing on machine learning, has been active in the area since 2013.

The company has launched sets of platforms using the techniques, which includes translation and voice research services.

“Experts in machine learning are flocking to the global IT giants, including Google,” a company official said. “Before competing with those machine learning market leaders, we are pushing to reinforce our manpower and expand the infrastructure for big data.”