Enhancing taxonomy-based extraction: Leveraging information from online community platforms for digital skills demand identification in job ads

Article type: Research Article

Authors: Napierała, Joanna^*

Correspondence: [*] Corresponding author: Joanna Napierała, The European Centre for the Development of Vocational Training (CEDEFOP), Cedefop, Service Post, Europe 123, 570 01 Thermi, Greece. E-mail: [email protected].

Abstract: The rapid technological changes have revolutionised how we function, including how we search for work and what skills we need to be equipped with to perform the tasks at the workplace. As employers more often recruit using online job advertisements, their content becomes a natural source of information for analytical purposes on the skills demanded in the labour market, especially for analysing emerging skills like digital. There are still some challenges with the extraction of information from online content. However, the extraction improvements go hand in hand with new technological developments like natural language processing techniques. This article presents the experimental method of updating the classification of digital skills to keep it up to date for information extraction applied to online job advertisements. The evaluation proved this method successfully identified terms related to programming skills but failed to identify terms associated with artificial intelligence sufficiently. The latter is related to the fact that the AI field is among the fastest developing areas of technology advancement, and new terms (e.g. Chatgpt) always appear.

Keywords: Digital skills, skills extraction, online job advertisements, big data, skills intelligence

DOI: 10.3233/SJI-230110

Journal: Statistical Journal of the IAOS, vol. Pre-press, no. Pre-press, pp. 1-12, 2024