Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

JobHop: A Large-Scale Dataset of Career Trajectories

Iman Johary, Raphael Romero, Alexandru C. Mara, Tijl De Bie

TL;DR

JobHop tackles the scarcity of public, large-scale, real-world career-trajectory data by introducing an LLM-based pipeline to extract structured experiences from anonymized resumes and map them to ESCO codes, resulting in over $1.67$ million experiences across more than $361{,}000$ resumes. The authors demonstrate the dataset's utility through analyses of education effects, career breaks, tenure, and transitions within the Flemish labor market, and validate the ESCO normalization against a baselined approach with human-annotation support. Key contributions include the two-stage extraction-and-normalization pipeline, a public ESCO-labeled career-trajectory dataset, and insights into how degree attainment and career interruptions shape occupational pathways. The dataset and findings offer a foundation for career-path prediction, workforce planning, and data-driven policy making, with planned updates to incorporate advances in LLM technology.

Abstract

Understanding labor market dynamics is essential for policymakers, employers, and job seekers. However, comprehensive datasets that capture real-world career trajectories are scarce. In this paper, we introduce JobHop, a large-scale public dataset derived from anonymized resumes provided by VDAB, the public employment service in Flanders, Belgium. Utilizing Large Language Models (LLMs), we process unstructured resume data to extract structured career information, which is then normalized to standardized ESCO occupation codes using a multi-label classification model. This results in a rich dataset of over 1.67 million work experiences, extracted from and grouped into more than 361,000 user resumes and mapped to standardized ESCO occupation codes, offering valuable insights into real-world occupational transitions. This dataset enables diverse applications, such as analyzing labor market mobility, job stability, and the effects of career breaks on occupational transitions. It also supports career path prediction and other data-driven decision-making processes. To illustrate its potential, we explore key dataset characteristics, including job distributions, career breaks, and job transitions, demonstrating its value for advancing labor market research.

JobHop: A Large-Scale Dataset of Career Trajectories

TL;DR

JobHop tackles the scarcity of public, large-scale, real-world career-trajectory data by introducing an LLM-based pipeline to extract structured experiences from anonymized resumes and map them to ESCO codes, resulting in over million experiences across more than resumes. The authors demonstrate the dataset's utility through analyses of education effects, career breaks, tenure, and transitions within the Flemish labor market, and validate the ESCO normalization against a baselined approach with human-annotation support. Key contributions include the two-stage extraction-and-normalization pipeline, a public ESCO-labeled career-trajectory dataset, and insights into how degree attainment and career interruptions shape occupational pathways. The dataset and findings offer a foundation for career-path prediction, workforce planning, and data-driven policy making, with planned updates to incorporate advances in LLM technology.

Abstract

Understanding labor market dynamics is essential for policymakers, employers, and job seekers. However, comprehensive datasets that capture real-world career trajectories are scarce. In this paper, we introduce JobHop, a large-scale public dataset derived from anonymized resumes provided by VDAB, the public employment service in Flanders, Belgium. Utilizing Large Language Models (LLMs), we process unstructured resume data to extract structured career information, which is then normalized to standardized ESCO occupation codes using a multi-label classification model. This results in a rich dataset of over 1.67 million work experiences, extracted from and grouped into more than 361,000 user resumes and mapped to standardized ESCO occupation codes, offering valuable insights into real-world occupational transitions. This dataset enables diverse applications, such as analyzing labor market mobility, job stability, and the effects of career breaks on occupational transitions. It also supports career path prediction and other data-driven decision-making processes. To illustrate its potential, we explore key dataset characteristics, including job distributions, career breaks, and job transitions, demonstrating its value for advancing labor market research.
Paper Structure (18 sections, 7 figures, 6 tables)

This paper contains 18 sections, 7 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Labor Market Datasets
  4. Dataset Curation
  5. Named Entity Recognition
  6. Occupation Classification
  7. Dataset Curation
  8. Data Overview
  9. Extracting Experiences and Qualifications
  10. Data Normalization
  11. Validation
  12. Data Cleaning
  13. Data Analysis
  14. How does holding a tertiary degree influence the types of occupations individuals pursue?
  15. How do career breaks impact subsequent occupation opportunities, and does this effect differ for individuals with a tertiary degree?
  16. ...and 3 more sections

Figures (7)

  • Figure 1: Structured output format expected to be returned by the LLM.
  • Figure 2: High level diagram of how the function $f(\cdot)$ works. It takes a job title as input and returns the corresponding ESCO code.
  • Figure 3: Normalized ratio of occupation category distribution between individuals with and without tertiary degrees, weighted by years in each occupation.
  • Figure 4: Fifteen most frequent post-career break employment outcomes by educational attainment. The top panel represents resumes of individuals with a tertiary degree, and the bottom panel represents resumes of individuals without a tertiary degree.
  • Figure 5: Percentage distribution of ESCO groups following a career break. The left panel shows all resumes, the middle panel focuses on individuals with a tertiary degree, and the right panel displays individuals without a tertiary degree.
  • ...and 2 more figures