Same as T209224: Analyze effect of huwiki FlaggedRevs configuration change on problematic edits and new user retention but for the T210224: Revert FlaggedRevs changes on the Hungarian Wikipedia changes (deployed on 2020-12-10) instead of the original T121995: Switch FlaggedRevs on Hungarian Wikipedia to a "flagged protection" mode changes (deployed on 2018-04-09).
Description
Description
Status | Subtype | Assigned | Task | ||
---|---|---|---|---|---|
Open | None | T223892 [EPIC] Support Hungarian Wikipedia editor retention project | |||
Open | None | T325925 Analyze effect of huwiki FlaggedRevs configuration change on problematic edits and new user retention (round2) | |||
Open | None | T223900 Create ORES dataset for huwiki edits in the last two years or so |
Event Timeline
Comment Actions
An interesting tool Nettrom has shown me is CausalImpact which seems like a good choice for this kind of time series analysis. (The time series analysis for the previous round was done by @Samat using JDemetra+, but that relies on patterns in the time series itself for excluding calendar effects, while CausalImpact takes a more clever approach where it can also use patterns in similar but unrelated time series data.)