de Maricourt, Claire (2018) Online job offers analysis PFE - Project Graduation, ENSTA.

Available under License Creative Commons Attribution Non-commercial No Derivatives.



Online job offers analysis is a new challenge for national statistical services. Indeed it reflects labor market shape and enables a better comprehension of its evolution. French labor market statistical service (Dares) is interested in studying online job offers to estimate more precisely tightness on labor market (which is a comparison between job offers and unemployment). It requires to collect online job offers (by developing scraping tools for example) and to classify job offers according to official French job nomenclature (called “code ROME”). Job offers are classified according to their titles. Different approaches are being compared: Tf-Idf and cosine similarity, Jaro-Winkler similarity, Damerau-Levenshtein similarity.

Item Type:Thesis (PFE - Project Graduation)
Uncontrolled Keywords:Online job offers, labor market, employment service, web-scraping, ROME code, NLP, cosinus similarity, textual analysis, string matching, classification, Damerau-Levenshtein, Jaro-Winkler, Tf-Idf
Subjects:Information and Communication Sciences and Technologies
Mathematics and Applications
ID Code:7287
Deposited By:Claire De Maricourt
Deposited On:27 mars 2019 11:00
Dernière modification:27 mars 2019 11:00

Repository Staff Only: item control page