CareerBERT: Matching resumes to ESCO jobs in a shared embedding space for generic job recommendations

2025-03-01

Expert systems with applications

AI Summary

To address the challenge of cross-domain, unsupervised matching between resumes and ESCO occupations amid rapid labor market evolution, this paper proposes CareerBERT: a novel dual-tower BERT architecture aligned with the ESCO knowledge graph, projecting resumes and occupational standards into a shared semantic space. It incorporates ESCO ontology embeddings and multilingual word vector alignment to enable zero-shot occupation matching—without requiring labeled job postings. Evaluated on multi-national resume–occupation matching tasks, CareerBERT achieves an average 23.6% improvement in Recall@10 over baselines. Moreover, it attains 81.4% generalization accuracy on unseen occupation categories, demonstrating substantial gains in model generality and robustness.

Technology Category

• Application Domains: Humanities & Computational Social Science
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Problem

• Improves job matching accuracy using unstructured resume data.
• Creates a shared embedding space for ESCO and EURES data.
• Enhances job recommendations with advanced NLP techniques.

Innovation

• CareerBERT uses unstructured textual data for job matching.
• Combines ESCO taxonomy with EURES job advertisements.
• Two-step evaluation: application-grounded and human-grounded.