A new study by the International Labour Organisation shows that unemployment will rise by 2.5 million this year. Around half a billion people are working fewer hours than they would like to or are not getting enough access to paid work, it said.

Unemployment is a harrowing experience, and its social impact risks getting worse with part-time and freelance work becoming popular – if governments don’t take mitigating actions. A report by the Singapore Government noted the rise of ‘gig economy’ companies that take “virtually zero duty of care” for employees’ mental wellbeing.

Governments are turning to data and AI to stem unemployment and respond. We look at three examples of how countries are using data and AI to recommend jobs to unemployed citizens; clamp down on illegal hires; and identify emerging skills and industries.

1. Recommend jobs

Estonia is using machine learning to profile employment candidates and recommend types of jobs that they should apply for. Algorithms could predict which jobs are in danger of disappearing, and “give a recommendation of where you could go next”, the country’s Chief Data Officer Ott Velsberg told GovInsider. “We could, in theory, give a heads up to you and say that you should perhaps choose another one,” he said.

AI-powered job matching “actually has a higher success rate than we previously had”, he added. It looks at their work history and skills to match with employers. Around 72 per cent of candidates who joined new jobs through the AI system were still employed six months later, compared to 58 per cent of those advised by officials, Wired reported.

2. Monitor fraud and illegal hires

Singapore is using data to track the risk of employees and employers trying to break labour laws.

“Using advanced data science and machine learning on large, diverse sets of employment, business and transactional data, MOM developed agile analytics models to detect patterns and emerging risks for which early interventions in policy and regulations could be implemented,” a Ministry of Manpower (MOM) spokesperson told GovInsider.

Greater use of data has allowed the Ministry to become more efficient in tracking employment fraud. “With tighter controls, more people would try to circumvent the system. With better use of data, MOM has achieved significantly higher detection rates and greater enforcement presence in a resource lean manner. For one exercise, we saved about 20,000 man-hours but covered 650 more cases,” the Ministry said.

3. International migration and future skills

The United Nations Development Programme’s Regional Innovation Centre for Asia Pacific has used publicly available LinkedIn data to identify labour migration patterns to and from Thailand. The preliminary analysis shows how governments could include real-time data from online platforms and job boards to identify emerging trends on the future of work, wrote Shumin Liu, the centre’s Data and Impact Management Consultant.

It showed, for example, that Thailand attracted most people to jobs in international affairs, renewables, and sports. Meanwhile, defence, space and maritime lost the most jobs to employees migrating out of Thailand. Between 2015 and 2018, the biggest changes in labour migration came to and from Thailand were in the US, France and Myanmar. The LinkedIn data also showed that soft skills like time management, problem solving, negotiation and leadership are on the rise in Thailand.

The UN team plans to look at how it can supplement this analysis with other and more representative data. “We are thinking whether we can apply a similar approach to excavate data on the local professional networking platforms like LinkedIn and link it to official data in governments and the database from ILO, connecting the fragmented data residing in different systems,” Liu said.

Unemployment has a far-reaching impact on daily lives, including relationships, health and mental wellbeing. Governments will need to identify their impact early on and ensure employees have stability in their work.