Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enhancing Job Matching: Occupation, Skill and Qualification Linking with the ESCO and EQF taxonomies

Stylianos Saroglou, Konstantinos Diamantaras, Francesco Preta, Marina Delianidi, Apostolos Benisis, Christian Johannes Meyer

TL;DR

This work investigates how to map job vacancy text to European labor classifications by comparing sentence-level versus entity-level linking to ESCO and EQF. It introduces three novel datasets, builds an open-source classifier toolkit, and conducts extensive experiments across occupation, skill, and qualification extraction. The results show that sentence linking best captures occupations when full context is available, whereas entity linking excels for skills, with qualifications remaining inconclusive; surprisingly, generative LLMs did not yield immediate improvements. The study also reports state-of-the-art entity recognition performance on the Green Benchmark and discusses limitations, future directions, and the potential of hybrid retrieval approaches for labor market narratives.

Abstract

This study investigates the potential of language models to improve the classification of labor market information by linking job vacancy texts to two major European frameworks: the European Skills, Competences, Qualifications and Occupations (ESCO) taxonomy and the European Qualifications Framework (EQF). We examine and compare two prominent methodologies from the literature: Sentence Linking and Entity Linking. In support of ongoing research, we release an open-source tool, incorporating these two methodologies, designed to facilitate further work on labor classification and employment discourse. To move beyond surface-level skill extraction, we introduce two annotated datasets specifically aimed at evaluating how occupations and qualifications are represented within job vacancy texts. Additionally, we examine different ways to utilize generative large language models for this task. Our findings contribute to advancing the state of the art in job entity extraction and offer computational infrastructure for examining work, skills, and labor market narratives in a digitally mediated economy. Our code is made publicly available: https://github.com/tabiya-tech/tabiya-livelihoods-classifier

Enhancing Job Matching: Occupation, Skill and Qualification Linking with the ESCO and EQF taxonomies

TL;DR

This work investigates how to map job vacancy text to European labor classifications by comparing sentence-level versus entity-level linking to ESCO and EQF. It introduces three novel datasets, builds an open-source classifier toolkit, and conducts extensive experiments across occupation, skill, and qualification extraction. The results show that sentence linking best captures occupations when full context is available, whereas entity linking excels for skills, with qualifications remaining inconclusive; surprisingly, generative LLMs did not yield immediate improvements. The study also reports state-of-the-art entity recognition performance on the Green Benchmark and discusses limitations, future directions, and the potential of hybrid retrieval approaches for labor market narratives.

Abstract

This study investigates the potential of language models to improve the classification of labor market information by linking job vacancy texts to two major European frameworks: the European Skills, Competences, Qualifications and Occupations (ESCO) taxonomy and the European Qualifications Framework (EQF). We examine and compare two prominent methodologies from the literature: Sentence Linking and Entity Linking. In support of ongoing research, we release an open-source tool, incorporating these two methodologies, designed to facilitate further work on labor classification and employment discourse. To move beyond surface-level skill extraction, we introduce two annotated datasets specifically aimed at evaluating how occupations and qualifications are represented within job vacancy texts. Additionally, we examine different ways to utilize generative large language models for this task. Our findings contribute to advancing the state of the art in job entity extraction and offer computational infrastructure for examining work, skills, and labor market narratives in a digitally mediated economy. Our code is made publicly available: https://github.com/tabiya-tech/tabiya-livelihoods-classifier

Paper Structure

This paper contains 12 sections, 1 figure, 12 tables.

Table of Contents

  1. Introduction
  2. Datasets
  3. Reference Sets
  4. Evaluation Sets
  5. Training Sets
  6. Methodology #1 : Sentence Linking
  7. Methodology #2: Entity Linking
  8. Evaluation
  9. Experiments
  10. Methodologies comparison
  11. Transformer Decoder Integration
  12. Conclusion

Figures (1)

  • Figure 1: F1-Score results on the Green Benchmark test set. The results show the mean and standard deviation over three random seeds.