Files
Citation
Share
Date
Abstract
1.- This report establishes fundamental efforts for the immediate identification of reskilling and upskilling needs through a semi-supervised and supervised approach that integrates advanced language models, such as GPT-4o, and transformer models such as BETO and mBERT, along with human supervision. This method allowed for the creation and expansion of an adaptive Knowledge, Skills and Abilities (KSA) taxonomy, ensuring relevance to industry needs and achieving 78% classification accuracy. 2. The report presents the distribution of KSAs, highlighting the predominance of technical knowledge (67%) compared to practical skills (20%) and abilities (7%). It identifies the most important KSAs, emphasizing teamwork, communication, and technical expertise as the most in-demand competencies. 3. It highlights three main categories, based on one month of published job postings on multiple websites in Mexico: technical, commercial, and administrative roles. The most common technical positions include Mechanical Technicians and Quality Inspectors. The report maps 30 occupations encompassing 1,723 unique job titles, representing 40% of job postings. 4. Additionally, knowledge network clusters reveal subgroups of administrative knowledge (e.g. customer service) and technical knowledge (e.g. mechanical maintenance). Meanwhile, the skills network makes distinctions between operational skills (e.g. teamwork) and leadership skills (e.g. decision-making). Although smaller, the abilities network highlights respect and time management as key abilities. 5. Finally, the report identifies the distribution of 30 main roles in technical, commercial, and administrative categories, reflecting the demands of the sector and the emphasis on specialized competencies.