Ready to change careers? AI can help you with that


The typical Australian will change careers five to seven times in their working life, by some estimates. And this is likely to increase as new technologies automate work, production is moved overseas, and economic crises unfold.

The loss of jobs is not a new phenomenon – have you recently seen an elevator operator? – but the pace of change is accelerating, threatening to leave large numbers of workers unemployed and unemployable.

New technologies also create new jobs, but the skills they require do not always match old jobs. To successfully transition from one job to another, you need to make the most of your current skills and learn new ones, but these transitions can weaken if the gap between old and new skills is too large.

We’ve built a system for recommending career transitions, using machine learning to analyze over 8 million online job postings to see which changes are likely to be successful. Details are published in PLOS A.

Our system starts by measuring the similarities between the skills required by each occupation. For example, an accountant might become a financial analyst because the skills required are similar, but a speech-language pathologist might have more difficulty becoming a financial analyst because the skills are very different.

Next, we looked at a wide range of real-world career transitions to see which direction those transitions typically go: Accountants are more likely to become financial analysts than the other way around.

Finally, our system can recommend a potentially successful career change – and tell you what skills you might need to make it work.

Measure the similarity of occupations

Our system uses a measure economists call “revealed comparative advantage” (RCA) to identify the importance of an individual skill to a job, using online job postings from 2018.

The map below visualizes the similarity of the top 500 skills. Each marker represents an individual skill, colored according to one of 13 very similar skill groups.

(Dawson et al., PLOS One, 2021)

Above: The similarity between the Top 500 Skills in Australian Jobs 2018. Very similar skills cluster together.

Once we know how similar the different skills are, we can estimate how similar the different professions are based on the skills required. The figure below visualizes the similarity between Australian occupations in 2018.

Each marker indicates an individual profession and the colors represent the risk each profession faces due to automation over the next two decades (blue indicates low risk and red indicates high risk).

Visibly similar professions are grouped closely together, with medical and highly skilled professions facing the lowest risk of automation.

folder 20210718 23 14e8pl7(Dawson et al., PLOS One, 2021)

Above: The similarity between the trades, tinged by the risk of technological automation.

Map transitions

We then took our occupational similarity measure and combined it with a range of other labor market variables, such as employment levels and education requirements, to create our referral system. professional transition.

Our system uses machine learning techniques to “learn” real job transitions in the past and predict job movements in the future. Not only does it achieve high levels of precision (76%), but it also takes into account asymmetries between job transitions.

Performance is measured by how accurately the system predicts whether a transition has occurred, when applied to historical job transitions.

The full transitions map is big and complicated, but you can see how it works below in a small version that only includes transitions between 20 professions. On the map, the “source” occupation is represented on the horizontal axis and the “target” occupation on the vertical axis.

If you look at a given profession at the bottom of the map, the column of squares indicates the probability of moving from that profession to the one shown on the right.

The darker the square, the higher the probability of making the transition.

file 20210805 13 lm02k8(Dawson et al., PLOS One, 2021)

Above: A small piece of the transition map, with 20 professions.

Recommendations for use based on artificial intelligence

Sometimes a new career requires developing new skills, but what skills? Our system can help identify them. Let’s take a look at how it works for “household cleaners,” a profession where employment declined sharply during COVID-19 in Australia.

file 20210718 15775 1lodn02(Dawson et al., PLOS One, 2021)

Above: The job transition recommendation system for “household cleaners” – a “non-essential” profession that has seen significant declines during the COVID-19 outbreak in Australia.

First, we use the transition map to see which occupations it is easier for a household cleaner to switch to. Colors divide occupations according to their status during the COVID-19 crisis – blue occupations are “essential” jobs that can continue to function during the lockdown, and red ones are “non-essential”.

We identify the most recommended occupations, as seen on the right side of the flowchart (bottom half of the image), sorted in descending order by transition probability.

The width of each strip in the diagram indicates the number of openings available for each trade. The colors of the segments indicate whether demand has increased or decreased compared to the same period of 2019 (before COVID).

The first six transition recommendations concern all “non-essential” services, which unsurprisingly experienced a drop in demand. However, the seventh is that of “elderly and disabled caregivers,” which is classified as “essential” and whose demand increased significantly at the start of the COVID-19 period.

Since your chances of finding work are better if you move to a high-demand occupation, we select “elderly and disabled caregivers” as the target occupation for this example.

What skills to develop for new professions

Our system can also recommend skills that workers need to develop to increase their chances of a successful transition. We argue that a worker should invest in developing the skills that are most important for their new occupation and which are the most different from the skills they currently have.

For a “household cleaner”, the most recommended skills needed to transition to “elderly and disabled caregiver” are specialized skills in patient care, such as “patient hygiene assistance”.

On the other hand, there is less need to develop skills that are irrelevant or very similar to skills in your current profession. Skills such as ‘business analysis’ and ‘finance’ are of low importance to an ‘elderly and disabled caregiver’, so they should not be given priority.

Likewise, skills such as “ironing” and “laundry” are required for the new job, but it is likely that a “housekeeper” already has these skills (or can easily acquire them).

The benefit of smoother job transitions

While the future of work remains uncertain, change is inevitable. New technologies, economic crises and other factors will continue to alter labor demands, forcing workers to change jobs.

If work transitions go smoothly, there are significant productivity and fairness benefits for everyone. If the transitions are slow or fail, it will incur significant costs for individuals, the state and the individual. The methods and systems we propose here could greatly improve the achievement of these goals.

We thank Bledi Taska and Davor Miskulin of Burning Glass Technologies for generously providing the job posting data for this research and for their valuable comments. We also thank Stijn Broecke and other OECD colleagues for their continued contribution and guidance in developing this work.The conversation

Nik Dawson, Honorary Fellow, Sydney University of Technology; Marian-Andrei Rizoiu, Senior Lecturer in Computer Science, Sydney University of Technology, and Mary-Anne Williams, Michael J Crouch Chair in Innovation, UNSW.

This article is republished from The Conversation under a Creative Commons license. Read the original article.


Comments are closed.