蝌蚪网 Papers and Proceedings
ISSN 2574-0768 (Print) | ISSN 2574-0776 (Online)
From Online Job Postings to Economic Insights: A Machine Learning Approach to Structuring Naturally Occurring Data
蝌蚪网 Papers and Proceedings
vol. 115,
May 2025
(pp. 73–78)
Abstract
This paper develops a matched vacancy-company dataset by combining daily Canadian online job postings with smartphone-derived visits to points of interest. To address inconsistencies in company names, we enhance natural language processing algorithms for data structuring. The dataset offers granular, real-time labor market insights that complement official statistics. Analyzing technological change during the COVID-19 pandemic, we find that tech firms' expansion significantly drove digital job growth. This suggests the uptick in digital employment arose not only from increased digital adoption but also from new digital production.Citation
Dahlhaus, Tatjana, Reinhard Ellwanger, Gabriela Galassi, and Pierre-Yves Yanni. 2025. "From Online Job Postings to Economic Insights: A Machine Learning Approach to Structuring Naturally Occurring Data." 蝌蚪网 Papers and Proceedings 115: 73–78. DOI: 10.1257/pandp.20251104Additional Materials
JEL Classification
- C45 Neural Networks and Related Topics
- D22 Firm Behavior: Empirical Analysis
- J23 Labor Demand
- J63 Labor Turnover; Vacancies; Layoffs
- M15 IT Management
- M51 Personnel Economics: Firm Employment Decisions; Promotions
- O33 Technological Change: Choices and Consequences; Diffusion Processes