Skip to contents

Build a comprehensive profession-skill profile table

Source: R/extract.R
build_profession_skill_profile.Rd

Joins announcements, skills, ISCO classification, a master skills list, and co-occurrence data into a single profession-skill table. This reproduces the join logic from the profession-skill exploration script (08_ski_ann_stru.R).

Usage

build_profession_skill_profile(ann, ski, isco, skillist, cooc = NULL)

Arguments

ann

A data.table of announcements with at least general_id, idesco_level_4, gdate, edate.

ski

A data.table of skills with at least general_id and ESCOSKILL_LEVEL_3.

isco

A data.table of ISCO groups with at least code and preferredLabel.

skillist

A data.table master skills list with at least ESCOSKILL_LEVEL_3 (or escoskill_level_3) and diffusione.

cooc

Optional data.table co-occurrence edge list with from, to, weight. If provided, it is returned unmodified as an attribute of the output for downstream graph analysis.

Value

A data.table with columns preferredLabel, ESCOSKILL_LEVEL_3, diffusione, and annunci (number of unique announcements per profession-skill pair).

Details

Steps:

  1. Merge announcements with ISCO to get preferredLabel.

  2. Merge skills with announcements on general_id.

  3. Merge with the master skills list for diffusion metadata.

  4. Aggregate to profession-skill level with announcement counts.