Automatisk jobbmatchning

Examensarbete på grundnivå

Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.12380/300032
Download file(s):
File Description SizeFormat 
CSE 19-07 CPL Karlsson Rosengren.pdfCSE Karlsson Rosengren510.85 kBAdobe PDFThumbnail
View/Open
Full metadata record
DC FieldValueLanguage
dc.contributor.authorKarlsson, Martin-
dc.contributor.authorRosengren, Tobias-
dc.contributor.departmentChalmers tekniska högskola / Institutionen för data och informationstekniksv
dc.date.accessioned2019-07-10T13:20:23Z-
dc.date.available2019-07-10T13:20:23Z-
dc.date.issued2019sv
dc.date.submitted2019
dc.identifier.urihttps://hdl.handle.net/20.500.12380/300032-
dc.description.abstractIn order to facilitate the workflow of recruiters who work daily with finding suitable assignments for their consultants, a system was created that can automatically match these consultants against different job listings. Creating such a system was a complex task since job listings often consist of unstructured text which is difficult to interpret for a computer. To deal with this problem, Natural Language Processing was applied, which is a methodology for managing unstructured text using artificial intelligence and various algorithms. The completed system delivered several positive results, in addition to these some misleading ones were also produced. The system’s results are recommended to be used as guidelines to pursue for a recruiter when a suitable candidate is to be selected for an assignment. There is much room for development to improve the system’s reliability and to reduce the likelihood of misleading results.sv
dc.language.isoswesv
dc.setspec.uppsokTechnology
dc.subjectNLPsv
dc.subjectComputational Linguisticssv
dc.subjectJob Matchingsv
dc.titleAutomatisk jobbmatchningsv
dc.type.degreeExamensarbete på grundnivåsv
dc.type.uppsokM
dc.contributor.examinerLundin, Peter-
dc.identifier.coursecodeLMTX38sv
Collection:Examensarbeten på grundnivå // Basic Level Theses



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.