ALL libraries (COBIB.SI union bibliographic/catalogue database)
  • Interactive skill based labor market mechanics and dynamics analysis system using machine learning and big data [Elektronski vir]
    Mršić, Leo ; Jerković, Hrvoje ; Balković, Mislav
    Interest in talent recognition, talent recruitment, education and labor mobility has been on the rise in last years. Business sector is changing its human resources (HR) policies globally, ... changingways in policies and practices related to employee management and employer branding. The process of talent recognition and/or search was mostly manual work, often carried by professional agencies or HR officer in organization. In todayʼs challenging labor market environment, this process is inefficient and slow with and often limited in success. In this paper we are focused on skills, education and lifelong learning domain which has an important role in the 10 priorities of the European Commission (EC) 2014-2019. Our starting point was to look for machine learning and big data techniques to support the policy makers and analysts in reducing mismatch between jobs and skills at regional level in the European Union (EU) through the use of data. Our research includesmassive semi structured resume dataset (50+ million documents) combined with several official statistical surveys. We were able to leverage the advances in machine learning and big data to automate resume/skill classification and to improve productivity in skill-based labor market mechanics and dynamics analysis. This paper proposes a model of extracting important information after resumes are being partially classified, classify skills using European multilingual classification of Skills, Competences, Qualifications and Occupations (ESCO) by skill pillar matching and use modern interactive visualization tools to gain smart insights powered by geospatial and time-series analysis. Research goals are to point attention towards data science and machine learning and its usage in labor and educational market mechanics and dynamics.
    Type of material - conference contribution
    Publish date - 2020
    Language - english
    COBISS.SI-ID - 72157699
    DOI