Page MenuHomePhabricator

Develop a ML-based service to predict reverts on Wikipedia(s)
Open, In Progress, HighPublic

Description

The Research team in collaboration with the ML-Platform team are creating a new service to help patrollers to detect revisions that might be reverted.

Requirments:

  • One single model for all Wikipedia languages. (use wiki_db as parameter)
  • Model should be primarly language agnostic (Check the subtasks)
  • Model will be able to run for single revisions or batches
  • Model should be able to run in Lift Wing

Please follow the progress of this project on the related tasks.

Event Timeline

Reedy renamed this task from Develop a ML-based service to predict reverts on Wikipedia(s) to Develop a ML-based service to predict reverts on Wikipedia(s).Aug 2 2022, 12:47 PM
diego changed the task status from Open to In Progress.Aug 2 2022, 12:53 PM
diego claimed this task.
diego triaged this task as High priority.
diego added projects: Research, Epic.
diego updated the task description. (Show Details)
diego added subscribers: calbon, AikoChou, MunizaA.
diego added a subscriber: leila.

It has been decided to focus on knowledge integrity risks from two categories of our taxonomy:

  • Content: prevalence and response to vandalism (using data generated from T314384)
  • Community: capacity (shortage of resurces in content moderation | admin burnout), governance (barriers to adminship rights) and demographics (geographical diversity of editors/readers)

It has been decided to focus on knowledge integrity risks from two categories of our taxonomy:

I think this comment shouldn't go on this task.

For the records here a snippet (by @achou) to try the models from the WMF's cluster

Language-Agnostic:

curl "https://inference.svc.codfw.wmnet:30443/v1/models/revert-risk-model:predict" -d @input.json -H "Host: revert-risk-model.experimental.wikimedia.org" --http1.1 -k

Multilingual:

curl "https://inference-staging.svc.codfw.wmnet:30443/v1/models/revert-risk-model:predict" -d @input.json -H "Host: revert-risk-model.experimental.wikimedia.org" --http1.1 -k

An example for input.json: { "lang": "ru", "rev_id": 123855516 }