Answers are grounded in this paper and cite their sources.
This thesis develops a strategy to identify digital skill demands in software engineering, suggesting improvements to labor classification systems.
The rapid pace of technological change, digitalization, and the emergence of new job roles challenge traditional labor market classification systems such as ESCO (European Skills, Competences, Qualifications and Occupations) and ISCO (International Standard Classification of Occupations). While these taxonomies provide valuable structure, they struggle to capture emerging competences in real time. This thesis develops a multi-layered enrichment strategy that integrates community-driven data and structured datasets to enhance the relevance of international classifications. Three frameworks are central to this work. First, ESCOPlus preprocessing was applied to extract and analyze skills from Stack Overflow, categorizing them into ESCO-aligned and non-ESCO skills. Second, ISCOccuPlus was developed to align ONET occupational descriptions with ISCO categories, ensuring that ISCO reflects new roles in ICT. Third, ISCOskillsPlus was designed to enrich ISCO competences with detailed O*NET skill examples by applying association rule mining, cosine similarity, and Jaccard similarity. The data pipeline combined API extraction, large-scale web scraping of Stack Overflow profiles, and the AUTH Skillab Tracker API. This hybrid approach produced a robust dataset that captured both established and emerging skills. The analysis revealed significant gaps between ESCO and real-world practice, with non-ESCO skills such as ReactJS, Docker, and Kubernetes underscoring the lag in official taxonomies. The contributions of this thesis are threefold: methodological, through the integration of hybrid similarity and association rule mining techniques; practical, by generating enriched datasets and mappings between O*NET and ISCO; and conceptual, by demonstrating how bottom-up evidence from developer communities can complement top-down classifications. Ultimately, this research provides a blueprint for continuous taxonomy enrichment, supporting labor market transparency, curriculum design, and workforce mobility in an era of rapid digital transformation.
No takes yet. Share an insight, caveat, or question.
Ιωσηφίνα Ε. Μαράκη (2025) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: