Skill-driven recommendations for job transition pathways

Abstract

Job security can never be taken for granted, especially in times of rapid, widespread and unexpected social and economic change. These changes can force workers to transition to new jobs. This may be because new technologies emerge or production is moved abroad. Perhaps it is a global crisis, such as COVID-19, which shutters industries and displaces labor en masse. Regardless of the impetus, people are faced with the challenge of moving between jobs to find new work. Successful transitions typically occur when workers leverage their existing skills in the new occupation. Here, we propose a novel method to measure the similarity between occupations using their underlying skills. We then build a recommender system for identifying optimal transition pathways between occupations using job advertisements (ads) data and a longitudinal household survey. Our results show that not only can we accurately predict occupational transitions (Accuracy = 76%), but we account for the asymmetric difficulties of moving between jobs (it is easier to move in one direction than the other). We also build an early warning indicator for new technology adoption (showcasing Artificial Intelligence), a major driver of rising job transitions. By using real-time data, our systems can respond to labor demand shifts as they occur (such as those caused by COVID-19). They can be leveraged by policy-makers, educators, and job seekers who are forced to confront the often distressing challenges of finding new jobs.

Artificial intelligence and employment

“… It misses vacancies that are not posted online. Carnevale, Jayasundera and Repnikov (2014 [13]) compare vacancies from survey data according to the Job Openings and Labor Turnover Survey (JOLTS) from the US Bureau of Labour Statistics, a representative survey of 16 000 US businesses, with BGT data for 2013. They find that roughly 70% of vacancies were posted online, with vacancies requiring a college degree significantly more likely to be posted online compared to jobs with lower education requirements.…”

“…peripheral vision. 13 The correlation matrix between applications and abilities is then calculated as the share of respondents who thought that a given AI application could be used for a given ability. These abilities are subsequently linked to occupations using the O*NET database.…”

“…(Felten, Raj and Seamans, 2021[35]). 13 The background of the gig workers is not known and so they may not necessarily be AI experts. This could be a potential weakness of this indicator.…”

Artificial Intelligence and Employment: New Cross-Country Evidence

“…Dawson et al ( 2021 ) develop the skills-space or skills-similarity indicator. This approach defines two skills as similar if they often occur together in BGT job postings and are both simultaneously important for the job posting.…”

“…For example, TensorFlow is a deep learning framework introduced in 2016. Many job postings now list it as a requirement without additionally specifying “deep learning” (Dawson et al, 2021 ).…”