Scholarlink

The recruitment process in IT companies is increasingly informed by standardized taxonomies such as European Skills, Competences, Qualifications, and Occupations (ESCO), which aim to structure and harmonize skill definitions across Europe. However, to remain effective in the rapidly evolving digital landscape, these taxonomies require continuous updates—particularly in areas like programming languages, frameworks, and specialized software development—a gap this study aims to address. We introduce ESCOPlus, a non-proprietary and open-source framework that enhances ESCO by integrating user-generated content from Stack Overflow (SO). Our approach extracts digital skills from the tags used by SO users in the European Union (EU) between 2020 and 2024, analyzes their co-occurrence with the ESCO taxonomy, and applies association rule mining to uncover strong relationships. Additionally, textual similarity measures identify alternative labels (closely related names to existing ESCO skills) to improve taxonomy completeness. As a final outcome, the ESCOPlus framework extended the taxonomy with 44 new digital skills. The extension of the ESCO taxonomy introduces a dynamic, data-driven methodology for taxonomy enrichment, enhancing skill classification in recruitment, HR analytics, and curriculum design to better align with evolving industry needs.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

ESCOPlus: A Framework for Enriching the ESCO Taxonomy with Digital Skills from Stack Overflow

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Dimitrios Christos Kavargyris,
  • Konstantinos Georgiou,
  • Iosifina Maraki,
  • Nikolaos Mittas,
  • Lefteris Angelis

摘要

The recruitment process in IT companies is increasingly informed by standardized taxonomies such as European Skills, Competences, Qualifications, and Occupations (ESCO), which aim to structure and harmonize skill definitions across Europe. However, to remain effective in the rapidly evolving digital landscape, these taxonomies require continuous updates—particularly in areas like programming languages, frameworks, and specialized software development—a gap this study aims to address. We introduce ESCOPlus, a non-proprietary and open-source framework that enhances ESCO by integrating user-generated content from Stack Overflow (SO). Our approach extracts digital skills from the tags used by SO users in the European Union (EU) between 2020 and 2024, analyzes their co-occurrence with the ESCO taxonomy, and applies association rule mining to uncover strong relationships. Additionally, textual similarity measures identify alternative labels (closely related names to existing ESCO skills) to improve taxonomy completeness. As a final outcome, the ESCOPlus framework extended the taxonomy with 44 new digital skills. The extension of the ESCO taxonomy introduces a dynamic, data-driven methodology for taxonomy enrichment, enhancing skill classification in recruitment, HR analytics, and curriculum design to better align with evolving industry needs.