In this blog series, international experts on the topic of analyzing online job advertisements present their work. The presentations were given at the 9th OJA Forum, which is jointly organized twice a year by the Federal Institute for Vocational Education and Training and the Bertelsmann Stiftung. Topics at the 9th OJA Forum included the further development of existing taxonomies. Taxonomies are systematic classification systems used to organize information according to specific criteria – for example, to classify occupations, skills, or qualifications. Another key topic was how to link training content with labor market requirements. In this video blog series, the presentations will be published one after the other. The fifth talk is by Julian Rosenberger from the University of Regensburg. 

Beyond Ad Matching: Using CareerBERT to Map Resumes to ESCO Job Categories via Semantic  

This talk introduces CareerBERT, a novel approach leveraging BERT embeddings to match unstructured resume data directly to standardized ESCO job classifications, moving beyond traditional matching to specific advertisements. We will discuss the creation of a shared semantic space using a hybrid corpus of ESCO taxonomy and EURES job ad data, detail the model architecture, and present evaluation results including quantitative metrics and qualitative feedback from HR experts. Discover how this method can enhance career guidance by providing broader, semantically relevant job recommendations based on the holistic content of a resume. 

 

 

Here you can find all videos from this series:

9th OJA Forum (Part 1) – What to add

9th OJA Forum (Part 2) – ESCO development

9th OJA Forum (Part 3) – Building and Evolving Green and AI Skill Taxonomies

9th OJA Forum (Part 4) – App to provide labor market guidance to students

9th OJA Forum (Part 5) – CareerBERT: Mapping Resumes to ESCO Jobs