Bridging Skill Gaps: Combining Generative and Symbolic AI for Personalized Lifelong Learning Pathways

Auteurs

Pruski C., da Costa Pereira C., Da Silveira M., Marconi G., Gallais M., Tettamanzi A.

Référence

Lecture Notes in Computer Science, vol. 15882 LNAI, pp. 19-27, 2025

Description

In today’s evolving job market, lifelong learning is essential for maintaining competitiveness. Workers must adapt to changing roles by identifying skill gaps and finding relevant training, yet this remains a challenge. This paper introduces a novel two-step approach combining generative and symbolic AI to support personalized lifelong learning. Large Language Models (LLMs) annotate résumés, job descriptions, and training offers with standardized vocabulary. A logic-based framework then detects missing skills and recommends appropriate training. Validated through real-world data, this method effectively supports skill gap analysis and tailored learning strategies.

Lien

doi:10.1007/978-3-031-98465-5_3