Abstract

Jobseekers get in touch with a large number of job offers on their way to re-employment. It is not always easy to distinguish which job offers are the most suitable. We have generated a data set that contains information on which person has taken up a particular job. This data is used to build a model based on machine learning methods (e.g. Extreme Gradient Boosting or Support Vector Machines). The aim is to develop a pre-selection of job offers for job seekers by calculating the probability that an application for a specific job will be successful. This means that a ranking of the job offers is created. In practice, this is intended to support the work of caseworkers.