Search papers, labs, and topics across Lattice.
The authors introduce UniSkill, a new dataset for matching university curricula to professional competencies, addressing a gap in publicly available resources for skill extraction and recommendation systems. The dataset includes both manually annotated and synthetically generated pairs of graduate-level university courses and skills from the ESCO taxonomy, focusing on Systems Analysts and Management/Organization Analyst occupations. They train and evaluate BERT models on this dataset, achieving an 87% F1-score, demonstrating the feasibility of automated course-to-skill matching.
Bridging the education-to-employment gap just got easier: UniSkill offers a novel dataset for precisely matching university courses to professional skills, enabling better curriculum design and personalized career guidance.
Skill extraction and recommendation systems have been studied from recruiter, applicant, and education perspectives. While AI applications in job advertisements have received broad attention, deficiencies in the instructed skills side remain a challenge. In this work, we address the scarcity of publicly available datasets by releasing both manually annotated and synthetic datasets of skills from the European Skills, Competences, Qualifications and Occupations (ESCO) taxonomy and university course pairs and publishing corresponding annotation guidelines. Specifically, we match graduate-level university courses with skills from the Systems Analysts and Management and Organization Analyst ESCO occupation groups at two granularities: course title with a skill, and course sentence with a skill. We train language models on this dataset to serve as a baseline for retrieval and recommendation systems for course-to-skill and skill-to-course matching. We evaluate the models on a portion of the annotated data. Our BERT model achieves 87% F1-score, showing that course and skill matching is a feasible task.
