Semantic Assistance Services for Career Integration and Personal Competence Development (SABINE)


Continuously, employers are complaining about a lack of qualified employees. At the same time, one can notice that a lot of qualified employees are unemployed, the education of employees is not adequate for the current demand, or promotion prospects are affected by misdirected in-house education. Demand and availability of employees need to be coordinated with the continuing education offered by employers. Thus, it is necessary that computer systems in recruitment agencies, personnel services, as well as human resources departments are able to interact. Indeed there are several existing systems for each of those services, but they are all isolated.

The SABINE project (German: “Semantische Assistenzdienste für die berufliche Integration und Persönliche Kompetenzentwicklung”) develops methods to interlink the databases of the different systems by means of semantic methods. Therefore, the project enables employers to better find qualified employees, and employees can better find suitable jobs. Semantic methods are currently based on manually developed ontologies. Developing and maintaining ontologies is a very complex task. Maintenance is e.g. necessary when new professions or qualifications emerge. Furthermore, such ontologies usually describe a specific domain, e.g. IT Consulting, which hurts their broad applicability.


The UKP Lab's contribution will be in methods which extract semantic knowledge from domain-independent sources like Wikipedia by means of statistical text analysis. The methods are fully automated, i.e. they can extract a remarkable amount of lexical-semantic information without manual intervention. They can be used stand-alone if there is no manually developed ontology for a certain domain, or as an extension if an existing ontology does not provide sufficient coverage of the relevant concepts. This way, for example semantically related CV profiles of prospective employees and job descriptions of prospective employers can be matched effectively and efficiently.


What to be? – Electronic Career Guidance Based on Semantic Relatedness

Iryna Gurevych, Christof Müller, Torsten Zesch

In:Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics, p. 1032--1039, Association for Computational Linguistics, Juni 2007. PDF

Information Extraction with the Darmstadt Knowledge Processing Software Repository (Extended Abstract)

Iryna Gurevych, Mark-Christoph Müller

In: Proceedings of the Workshop on Linguistic Processing Pipelines, Darmstadt, Germany, July 10, 2008. PDF


More details about our partners can be found at


  • Richard Eckart de Castilho, Project Coordinator
  • Prof. Dr. Iryna Gurevych, Principal Investigator
  • Benjamin Herbert, Doctoral Researcher


The project is funded by the German Ministry of Education and Research (BMBF)

Further information on SABINE: