Abstract

Online job portals collecting web vacancies have become important media for job demand and supply matching. They also represent a growing research area for the application of analytical methods to study the labour market using innovative data sources. Both the Knowledge Discovery in Databases approach and mixed supervised and unsupervised text mining approaches were typically applied to retrieve occupations associated with each web vacancy (ISCO classification up to level 4) and related skills. In the present paper we apply this method to a population of online web vacancies collected for three countries (Italy, UK and Germany) collected over a quarter in 2019, within an international project, to demonstrate the potentiality of informative power of such approach that can be considered as promising strategy providing effective support for decision making of several stakeholders such as government organizations, analysts, and recruitment agencies, as they allow for timely and fine-grained representations of complex labour market dynamics, in terms of trends, occupations, and skills. Finally, problems of representativeness that affect online vacancies are briefly discussed and possible approaches are proposed.

Keywords: Job vacancies; Scraping; Big data; Job classification