Skip to Main content Skip to Navigation
Conference papers

Language Modelling for Collaborative Filtering: Application to Job Applicant Matching

Abstract : This paper addresses a collaborative retrieval problem , the recommendation of job ads to applicants. Specifically, two proprietary databases are considered. The first one focuses on the context of unskilled low-paid jobs/applicants; the second one focuses on highly qualified jobs/applicants. Each database includes the job ads and applicant resumes together with the collaborative filtering data recording the applicant clicks on job ads. The proposed approach, called LAJAM, focuses on the semi-cold start recommendation problem of recommending new job ads to known applicants. This setting is relevant to the temporary job sector, of increasing importance for current job markets. LAJAM learns a continuous language model on the job ad space, trained to comply with the collaborative filtering metrics. This language model, implemented as a neural net, can flexibly take into account heterogeneous additional information, e.g. related to the posting time and geolocation of the job ads. The merits of the LAJAM approach are demonstrated comparatively to the state of the art on the thoroughly studied CiteULike database. The comparison of the public CiteULike database with the proprietary databases sheds some light on the specific difficulties of the job applicant matching problem.
Document type :
Conference papers
Complete list of metadata

Cited literature [40 references]  Display  Hide  Download
Contributor : Thomas Schmitt Connect in order to contact the contributor
Submitted on : Friday, December 8, 2017 - 2:56:27 PM
Last modification on : Thursday, July 8, 2021 - 3:50:43 AM


Files produced by the author(s)


  • HAL Id : hal-01659543, version 1


Thomas Schmitt, François Gonard, Philippe Caillou, Michèle Sebag. Language Modelling for Collaborative Filtering: Application to Job Applicant Matching. ICTAI 2017 - 29th IEEE International Conference on Tools with Artificial Intelligence, Nov 2017, Boston, United States. pp.1-8. ⟨hal-01659543⟩



Record views


Files downloads