•  
  •  
 

University of South Florida M3 Center Publishing

Abstract

Matching job candidates with job offerings is one of the most important business tasks and is crucial to the success of a company. But there is not much knowledge available about the quality of matchings processed automatically by software. With a specifically developed scoring system it becomes possible to make a statement about the quality of the matching results generated by three different tools, i.e., Textkernel, Joinvision and Sovren. A series of resumes is being matched against two concrete open job positions, one by Google and one by the University of Zurich. The results are then compared in detail with the human based assessment made by the authors. For the Post-Doctoral Researcher position at the University of Zurich the scoring results in general were weaker than for the Software Engineer position at Google. We found out that the success of a good matching depends mainly on the parsing of the CVs. The quality of CV information is depending on how it is structured and what the specific candidate’s experience is. The different tools showed that the ranking of candidates is dependent on the number of keyword matches. In particular for the job offer at Google, the available CVs obviously included suitable candidates. Textkernel and Sovren were capable to parse the CVs and job description correctly and therefore achieved good results, whereas Joinvision failed to extract key information and consequently dropped to the last place in the ranking.

DOI

https://www.doi.org/10.5038/9781955833035

Recommended Citation

Buttiker, F., Roth, S., Steinacher, T., & Hanne, T. (2021). Comparative analysis of tools for matching work-related skill profiles with CV data and other unstructured data. In C. Cobanoglu, & V. Della Corte (Eds.), Advances in global services and retail management (pp. 1–14). USF M3 Publishing. https://www.doi.org/10.5038/9781955833035

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.