Keywords: education, skill, job, NLP, machine learning, zero-shot learning
Project coordinator: Prof. Conny Wunsch
CeDA collaborator: Rodrigo C. G. Pena

Project objectives

  • To extract formal education required by jobs from original job ad text using modern Natural Language Processing (NLP) tools.
  • To assess and improve upon the skills extracted from a commercial job listings API platform.
  • To adjust education and skill requirements to the Swiss context.

This CeDA collaboration is related to the Job Analysis To Track Skill Demand Evolution (JADE):

The digital transformation causes substantial shifts in the tasks associated with specific jobs and the skills and competencies demanded by employers. The extent of possible disparities this will create on the labour market will crucially depend on how labour supply responds to these developments. Moreover, digitalisation disrupts entire management systems, creating both new opportunities, and challenges for the formation and design of employment relationships. JADE is part of a project funded by the Swiss National Science Foundation within the National Research Program “Digital Transformation” (NRP77). In this project, we investigate how the digital transformation changes employment relationships and the matching of labour supply and demand in Switzerland. In JADE, we use large and regularly updated representative data on job vacancies posted in Switzerland to uncover the skills and competencies demanded by employers and analyse how they change over time. This information will serve as input for studying how firms and workers respond to these changes.

We predict the education requirements from the job description using an approach called “zero-shot learning”. Some team of researchers trains a large neural network on a variety of languages and labels, teaching the network to identify what a text is about. The model is made available online to other researches, who can then adapt it, give it new text and a new set of labels (both of which the model has not seen before), and the model ranks which labels match the best with the text. Here you can see a demo of this strategy at work.

Analysis and Visualization

If we present the ad for a cook job at a restaurant, along with a set of potential education requirements to the zero-shot learning machine, we have the following example output.


Figure 1. With these lines of code, a text is input and a ranking of education requirement labels is output. For a job as a cook at a restaurant, education as "Koch/Köchin EFZ" is the top ranked among the candidate requirements.