Engineering job ads corpus

Title:

Engineering job ads corpus

Creator:

Cardenas Acosta, Ronald, Bello Medina, Kevin, Coronado, Alberto, and Villota, Elizabeth

Contributor:

National University of Engineering, Peru@@UNI-INIFIM-2015-07@@Peruvian labor market analysis@@ownFunds@@

Publisher:

National University of Engineering, Peru

Identifier:

http://hdl.handle.net/11234/1-2673

Subject:

job-advertisement, PoS tagging, and text corpora

Type:

text and corpus

Description:

The corpus presented consists of job ads in Spanish related to Engineering positions in Peru. The documents were preprocessed and annotated for POS tagging, NER, and topic modeling tasks. The corpus is divided in two components: - POS tagging/ NER training data: Consisting of 800 job ads, each one tokenized and manually annotated with POS tag information (EAGLE format) and Entity Label in BIO format. - Topic modeling training data: containing 9000 documents stripped from stopwords. Comes in two formats: * Whole text documents: containing all the information originally posted in the ad. * Extracted chunks documents: containing chunks extracted by custom NER models (expected skills, tasks to perform, and preferred major), as described in Improving Topic Coherence Using Entity Extraction Denoising (to appear)

Language:

Spanish

Rights:

Creative Commons - Attribution 4.0 International (CC BY 4.0)
http://creativecommons.org/licenses/by/4.0/
PUB

Source:

https://github.com/ronaldahmed/labor-market-demand-analysis

Harvested from:

LINDAT/CLARIAH-CZ repository

Metadata only:

false

Date:

2016