Non-article focused damage detection
Open, LowestPublic
Actions

Assigned To

None

Authored By

	Halfak
	Jan 26 2017, 6:49 PM

Description

Build an AI that detects damage specific to non-article namespaces.

Wiki thing it helps with:

Wrong entries in wiki-projects, possible bad-faith or good-faith

Things that might help us get this AI built:

What are pain points among help deskers, etc.?
Train on reverted edits outside of main namespace.

Event Timeline

Halfak created this task.Jan 26 2017, 6:49 PM

Restricted Application added a subscriber: Aklapper. · View Herald TranscriptJan 26 2017, 6:49 PM

Note that we currently train models on a random sample of edits. However, we could also train models on specific subsets. Project edits are rare in a pure random sample of edits.

Halfak triaged this task as Lowest priority.Feb 2 2017, 3:37 PM

Harej moved this task from Ideas to Epic on the Machine-Learning-Team board.Apr 3 2019, 5:03 AM

• ACraze moved this task from Epic to Backlog/Revscoring on the Machine-Learning-Team board.Jan 19 2021, 8:43 PM

Non-article focused damage detection Open, LowestPublicActions

Description

Event Timeline

Non-article focused damage detection
Open, LowestPublic
Actions