Page MenuHomePhabricator

generate occupations metrics
Open, HighPublic

Event Timeline

Use cases:

  • I want to be able to compare high level occupation trends between different countries, by grouping subsets of occupation categories like football players, dancers as sports people
  • I want to choose specific women based on high level occupation fields like Women scientists that should automatically encompass researchers and other subcategories so that I don’t have to write Sparql queries. Also, health care professionals (optometrist, audiologists, nurses, doctors) and STEM category.
notconfusing moved this task from beta backlog (mvp) to beta in-progress on the Humaniki board.

STEPS to implement back-end generation:

  1. Create a test, manually comparing denelezh and humaniki the athlete type Q2066131 for the closest available dates. Denelezh:2020-09-21, Humaniki: 2020-11-30

image.png (82×941 px, 17 KB)

Occupation Total Total with gender Females % Females Gap Males % Males Others % Others
athlete 802,302 796,715 117,866 14.794 % 678,780 85.197 % 69 0.009 %

  1. update config and config reader to generate occupcation-parent with depth 1
  1. Update _get_dim_join_from_dim_prop so that occupation join table is a subqueried table with parent and occupation to specified depth. May require retooling of _get_dim_join_from_dim_prop
  1. Backfill the occupation metrics for humaniki already processed dumps (Deploy backend to production.)
  1. Front-end integration in advanced search only. Deploy frontend