Page MenuHomePhabricator

Fit models for revert prediction
Open, NormalPublic

Description

Fit logistic regression models predicting probability of revert including features for:

  • is newcomer
  • ORES is deployed
  • is anon
  • ... (all the other features we have in the models)

To do this we have to:

  1. Create automated badwords lists for the large-enough wikis that don't adopt ores for rcfilters.
  2. Build datasets including the ores features for each wiki that did or didn't adopt oress for rcfilters.
  3. Add features for newcomer, ores being deployed, and is_anon
  4. fit diff-in-diff models combining these features.

Event Timeline

Halfak created this task.Jun 3 2019, 4:07 PM
Groceryheist updated the task description. (Show Details)Jun 3 2019, 5:58 PM
Harej triaged this task as Normal priority.Jun 11 2019, 9:02 PM
Harej moved this task from Untriaged to New development on the Scoring-platform-team board.

I would like to work on this issue :)