Determining Column Numbers in Résumés with Clustering


Keskin Ş. R. , Balı Y., Orman G. K. , Daniş F. S. , Turhan S. N.

18th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, AIAI 2022, Hersonissos, Greece, 17 - 20 June 2022, vol.647 IFIP, pp.460-471 identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 647 IFIP
  • Doi Number: 10.1007/978-3-031-08337-2_38
  • City: Hersonissos
  • Country: Greece
  • Page Numbers: pp.460-471
  • Keywords: DBSCAN, Information Extraction, K-means, Résumé Parse

Abstract

In the recruitment process, the workload of manual résumé reviews is quite time consuming for the recruiters. This review process can benefit from Artificial Intelligent-aided intelligent systems to extract the actual meaning within the résumés and structure their forms. However, writing résumés has no standards, and the personalized structure of each received résumé makes this task highly challenging. This work is dedicated to tackling a part of this issue on structuring résumés. More specifically, we firstly focus on finding the column number of any résumé since once the main parts of the résumé are separated, the subdivisions can easily be analysed. This study, thus, formalizes the problem of finding columns of a résumé as a clustering problem. The experiments are performed on a data set of custom Turkish résumés having up to two-columns, on which we apply two algorithms: K-means and Density-based spatial clustering of applications with noise. As a result of the experiments, we observe that an optimal cluster size relates strongly to the valid column number. Our method is not limited to résumés but can be applied to any unstructured textual data.