Zero Digital Competence: Training Needs through Data Mining towards an Innovative Digital Training System

Abstract

This study identifies teaching training needs through text mining to underpin an innovative digital training system that contributes to reducing the digital skills gap, comparing pre- and post-pandemic periods. A sequential qualitative design was employed, with semi-structured interviews with 21 specialists from five Spanish-speaking countries (pre-pandemic phase, 2018-2020) and focus groups with 6 specialists (post-pandemic phase, 2023). Data were analyzed using text mining techniques (bigram analysis) and qualitative microanalysis, applying theoretical saturation criteria and methodological triangulation. The pre-pandemic analysis revealed concerns focused on curricular aspects, assessment of learning, and university structure. The post-pandemic study showed a shift toward technologies supporting data, a humanizing approach to learning, and solving real-world problems using artificial intelligence. The most significant bigrams ("virtual-reality," "soft-skills," "artificial-intelligence") confirm this evolution. The findings support a Digital Training System based on learning communities mediated by AI, personalized trajectories, and a technical-humanistic balance, transcending traditional models to address the digital skills gap in teaching in an innovative and contextualized manner.